Search

Transparent objects are encountered frequently in our daily lives, yet recognizing them poses challenges for conventional vision sensors due to their unique material properties, not being well perceived from RGB or depth cameras. Overcoming this limitation, thermal infrared cameras have emerged as a solution, offering improved visibility and shape information for transparent objects. In this paper, we present TRansPose, the first large-scale multispectral dataset that combines stereo RGB-D, thermal infrared (TIR) images, and object poses to promote transparent object research. The dataset includes 99 transparent objects, encompassing 43 household items, 27 recyclable trashes, 29 chemical laboratory equivalents, and 12 non-transparent objects. It comprises a vast collection of 333,819 images and 4,000,056 annotations, providing instance-level segmentation masks, ground-truth poses, and completed depth information. The data was acquired using an FLIR A65 thermal infrared camera, two Intel RealSense L515 RGB-D cameras, and a Franka Emika Panda robot manipulator. Spanning 87 sequences, TRansPose covers various challenging real-life scenarios, including objects filled with water, diverse lighting conditions, heavy clutter, non-transparent or translucent containers, objects in plastic bags, and multi-stacked objects. Supplementary material can be accessed from the following link: https://sites.google.com/view/transpose-dataset. [ABSTRACT FROM AUTHOR]

A static world assumption is often used when considering the simultaneous localization and mapping (SLAM) problem. In reality, especially when long-term autonomy is the objective, this is not a valid assumption. This paper studies a scenario where landmarks can occupy multiple discrete positions at different points in time, where each possible position is added to a multi-hypothesis map representation. A selector-mixture distribution is introduced and used in the observation model. Each landmark position hypothesis is associated with one component in the mixture. The landmark movements are modeled by a discrete Markov chain and the Monte Carlo tree search algorithm is suggested to be used as component selector. The non-static environment model is further incorporated into the factor graph formulation of the SLAM problem and is solved by iterating between estimating discrete variables with a component selector and optimizing continuous variables with an efficient state-of-the-art nonlinear least squares SLAM solver. The proposed non-static SLAM system is validated in numerical simulation and with a publicly available dataset by showing that a non-static environment can successfully be navigated. [ABSTRACT FROM AUTHOR]

Vision-based estimation of the motion of a moving target is usually formulated as a bearing-only estimation problem where the visual measurement is modeled as a bearing vector. Although the bearing-only approach has been studied for decades, a fundamental limitation of this approach is that it requires extra lateral motion of the observer to enhance the target's observability. Unfortunately, the extra lateral motion conflicts with the desired motion of the observer in many tasks. It is well-known that, once a target has been detected in an image, a bounding box that surrounds the target can be obtained. Surprisingly, this common visual measurement especially its size information has not been well explored up to now. In this paper, we propose a new bearing-angle approach to estimate the motion of a target by modeling its image bounding box as bearing-angle measurements. Both theoretical analysis and experimental results show that this approach can significantly enhance the observability without relying on additional lateral motion of the observer. The benefit of the bearing-angle approach comes with no additional cost because a bounding box is a standard output of object detection algorithms. The approach simply exploits the information that has not been fully exploited in the past. No additional sensing devices or special detection algorithms are required. [ABSTRACT FROM AUTHOR]

Visual motion estimation is a well-studied challenge in autonomous navigation. Recent work has focused on addressing multimotion estimation in highly dynamic environments. These environments not only comprise multiple, complex motions but also tend to exhibit significant occlusion. Estimating third-party motions simultaneously with the sensor egomotion is difficult because an object's observed motion consists of both its true motion and the sensor motion. Most previous works in multimotion estimation simplify this problem by relying on appearance-based object detection or application-specific motion constraints. These approaches are effective in specific applications and environments but do not generalize well to the full multimotion estimation problem (MEP). This paper presents Multimotion Visual Odometry (MVO), a multimotion estimation pipeline that estimates the full SE (3) trajectory of every motion in the scene, including the sensor egomotion, without relying on appearance-based information. MVO extends the traditional visual odometry (VO) pipeline with multimotion segmentation and tracking techniques. It uses physically founded motion priors to extrapolate motions through temporary occlusions and identify the reappearance of motions through motion closure. Evaluations on real-world data from the Oxford Multimotion Dataset (OMD) and the KITTI Vision Benchmark Suite demonstrate that MVO achieves good estimation accuracy compared to similar approaches and is applicable to a variety of multimotion estimation challenges. [ABSTRACT FROM AUTHOR]

The modeling structures of rollers, mecanum wheels, and mecanum wheeled mobile robots presented in the literature use single contact force assumption. This assumption may give good results in a simulation environment; however, it is not strong enough to reflect reality. To make an improvement, a new aspect of mecanum wheel model is proposed in this study. The model takes the variable roller contact forces into account and investigates their effects on the performance of motion of a mecanum wheeled mobile robot. It uses all points on each roller's curved shape so that the slippage phenomena is also taken into consideration which makes it possible to get less position estimation errors in real-time operations. The modeling structure introduced aims to reflect reality both in simulation and real applications. A simulation environment is developed for this study. To make verification, an experimental setup including a four-mecanum-wheeled mobile robot, its mechanical and electrical hardware and software infrastructures, and a ground-truth system is designed and constructed. A Robot Operating System (ROS) based control system is created and integrated into the experimental system. Different types of reference trajectories including straight-line, square-shaped, Z-shaped, and wave(S)-shaped are used to test the performance of the model proposed in both simulation and experimental studies. The tests are also conducted using the model that involves single contact force assumption to make comparisons. The details of the variable contact forces model proposed, simulation environment developed, experimental setup built, simulation and experimental studies, their results, and comparisons are given in this paper. [ABSTRACT FROM AUTHOR]

Automation in agriculture is a growing area of research with fundamental societal importance as farmers are expected to produce more and better crop with fewer resources. A key enabling factor is robotic vision techniques allowing us to sense and then interact with the environment. A limiting factor for these robotic vision systems is their cross-domain performance, that is, their ability to operate in a large range of environments. In this paper, we propose the use of auxiliary tasks to enhance cross-domain performance without the need for extra data. We perform experiments using four datasets (two in a glasshouse and two in arable farmland) for four cross-domain evaluations. These experiments demonstrate the effectiveness of our auxiliary tasks to improve network generalisability. In glasshouse experiments, our approach improves the panoptic quality of things from 10.4 to 18.5 and in arable farmland from 16.0 to 27.5; where a score of 100 is the best. To further evaluate the generalisability of our approach, we perform an ablation study using the large Crop and Weed dataset (CAW) where we improve cross-domain performance (panoptic quality of things) from 12.8 to 30.6 for the CAW dataset to our novel WeedAI dataset, and 21.2 to 36.0 from CAW to the other arable farmland dataset. Although our proposed approaches considerably improve cross-domain performance we still do not generally outperform in-domain trained systems. This highlights the potential room for improvement in this area and the importance of cross-domain research for robotic vision systems. [ABSTRACT FROM AUTHOR]

Tactile servoing is an important technique because it enables robots to manipulate objects with precision and accuracy while adapting to changes in their environments in real-time. One approach for tactile servo control with high-resolution soft tactile sensors is to estimate the contact pose relative to an object surface using a convolutional neural network (CNN) for use as a feedback signal. In this paper, we investigate how the surface pose estimation model can be extended to include shear, and utilise these combined pose-and-shear models to develop a tactile robotic system that can be programmed for diverse non-prehensile manipulation tasks, such as object tracking, surface-following, single-arm object pushing and dual-arm object pushing. In doing this, two technical challenges had to be overcome. Firstly, the use of tactile data that includes shear-induced slippage can lead to error-prone estimates unsuitable for accurate control, and so we modified the CNN into a Gaussian-density neural network and used a discriminative Bayesian filter to improve the predictions with a state dynamics model that utilises the robot kinematics. Secondly, to achieve smooth robot motion in 3D space while interacting with objects, we used SE (3) velocity-based servo control, which required re-deriving the Bayesian filter update equations using Lie group theory, as many standard assumptions do not hold for state variables defined on non-Euclidean manifolds. In future, we believe that pose-and-shear-based tactile servoing will enable many object manipulation tasks and the fully-dexterous utilisation of multi-fingered tactile robot hands. [ABSTRACT FROM AUTHOR]

In this work we introduce a new type of human-inspired upper-limb prostheses. The Artificial Neuromuscular Prosthesis (ANP) imitates the human neuromuscular system in the sense of its compliance, backdrivability, natural motion, proprioceptive sensing, and kinesthetics. To realize this challenging goal, we introduce a novel human-inspired and simulation-based development paradigm to design the prosthesis mechatronics in correspondence to the human body. The ANP provides body awareness, contact awareness, and human-like contact response, realized via floating base rigid-body models, disturbance observers, and joint impedance control—concepts known from established state-of-the-art robotics. The ANP mechatronics is characterized by a four degrees of freedom (dof) torque-controlled human-like kinematics, a tendon-driven 2-dof wrist, and spatial orientation sensing at a weight of 1.7 kg (without hand and battery). The paper deals with the rigorous mathematical modeling, control, design and evaluation of this device type along initially defined requirements within a single prototype only. The proposed systemic and grasping capabilities are verified under laboratory conditions by an unimpaired user. Future work will increase the technology readiness level of the next generation device, where human studies with impaired users will be done. [ABSTRACT FROM AUTHOR]

Grasping is a key task for robots to interact with humans and the environment. Soft grippers have been widely studied and some have been applied in industry and daily life. Typical soft grippers face two challenges: lack of stiffness and insufficient adaptability to various objects. Inspired by the human hand, this paper proposes a soft-rigid hybrid pneumatic gripper composed of fingers with soft skin and rigid endoskeletons, and an active palm. Through different combinations of the four joints' locking states within the rigid endoskeleton, each finger obtains 9 different postures in its inflating state and 13 different postures in its deflating state, endowing the gripper with the capability of adapting to a wider variety of objects. Simultaneously, due to the endoskeletons, the lateral stiffness of the gripper is significantly enhanced (load-to-weight ratio∼7.5 for lateral grasping). We also propose a series of grasping strategies for grasping objects with different sizes and shapes to utilize the versatile configurations of the gripper. Experiments demonstrated that the gripper conformed well to the surfaces of cylindrical and prismatic objects and successfully grasped all tool items and shape items in the Yale–CMU–Berkeley object set. [ABSTRACT FROM AUTHOR]

Deformable linear objects (DLOs), such as rods, cables, and ropes, play important roles in daily life. However, manipulation of DLOs is challenging as large geometrically nonlinear deformations may occur during the manipulation process. This problem is made even more difficult as the different deformation modes (e.g., stretching, bending, and twisting) may result in elastic instabilities during manipulation. In this paper, we formulate a physics-guided data-driven method to solve a challenging manipulation task—accurately deploying a DLO (an elastic rod) onto a rigid substrate along various prescribed patterns. Our framework combines machine learning, scaling analysis, and physical simulations to develop a physics-based neural controller for deployment. We explore the complex interplay between the gravitational and elastic energies of the manipulated DLO and obtain a control method for DLO deployment that is robust against friction and material properties. Out of the numerous geometrical and material properties of the rod and substrate, we show that only three non-dimensional parameters are needed to describe the deployment process with physical analysis. Therefore, the essence of the controlling law for the manipulation task can be constructed with a low-dimensional model, drastically increasing the computation speed. The effectiveness of our optimal control scheme is shown through a comprehensive robotic case study comparing against a heuristic control method for deploying rods for a wide variety of patterns. In addition to this, we also showcase the practicality of our control scheme by having a robot accomplish challenging high-level tasks such as mimicking human handwriting, cable placement, and tying knots. [ABSTRACT FROM AUTHOR]

Signal temporal logic (STL) formulas have been widely used as a formal language to express complex robotic specifications, thanks to their rich expressiveness and explicit time semantics. Existing approaches for STL control synthesis suffer from limited scalability with respect to the task complexity and lack of robustness against the uncertainty, for example, external disturbances. In this paper, we study the online control synthesis problem for uncertain discrete-time systems subject to STL specifications. Different from existing techniques, we propose an approach based on STL, reachability analysis, and temporal logic trees. First, based on a real-time version of STL semantics, we develop the notion of tube-based temporal logic tree (tTLT) and its recursive (offline) construction algorithm. We show that the tTLT is an under-approximation of the STL formula, in the sense that a trajectory satisfying a tTLT also satisfies the corresponding STL formula. Then, an online control synthesis algorithm is designed using the constructed tTLT. It is shown that when the STL formula is robustly satisfiable and the initial state of the system belongs to the initial root node of the tTLT, it is guaranteed that the trajectory generated by the control synthesis algorithm satisfies the STL formula. We validate the effectiveness of the proposed approach by several simulation examples and further demonstrate its practical usability on a hardware experiment. These results show that our approach is able to handle complex STL formulas with long horizons and ensure the robustness against the disturbances, which is beyond the scope of the state-of-the-art STL control synthesis approaches. [ABSTRACT FROM AUTHOR]

In this paper, we introduce the concept of the Fluid Jacobian, which provides a description of the power transmission that operates between the fluid and mechanical domains in soft robotic systems. It can be understood as a generalization of the traditional kinematic Jacobian that relates the joint space torques and velocities to the task space forces and velocities of a robot. In a similar way, the Fluid Jacobian relates fluid pressure to task space forces and fluid flow to task space velocities. In addition, the Fluid Jacobian can also be regarded as a generalization of the piston cross-sectional area in a fluid-driven cylinder that extends to complex geometries and multiple dimensions. In the following, we present a theoretical derivation of this framework, focus on important special cases, and illustrate the meaning and practical applicability of the Fluid Jacobian in four brief examples. [ABSTRACT FROM AUTHOR]

This paper presents a novel method for efficiently solving a trajectory planning problem for swarm robotics in cluttered environments. Recent research has demonstrated high success rates in real-time local trajectory planning for swarm robotics in cluttered environments, but optimizing trajectories for each robot is still computationally expensive, with a computational complexity from O (k (n t , ε) n t 2) to O (k (n t , ε) n t 3) where n t is the number of parameters in the parameterized trajectory, ε is precision, and k (n t , ε) is the number of iterations with respect to n t and ε. Furthermore, the swarm is difficult to move as a group. To address this issue, we define and then construct the optimal virtual tube, which includes infinite optimal trajectories. Under certain conditions, any optimal trajectory in the optimal virtual tube can be expressed as a convex combination of a finite number of optimal trajectories, with a computational complexity of O (n t) . Afterward, a hierarchical approach including a planning method of the optimal virtual tube with minimizing energy and distributed model predictive control is proposed. In simulations and experiments, the proposed approach is validated and its effectiveness over other methods is demonstrated through comparison. [ABSTRACT FROM AUTHOR]

Robotic pick and place tasks are symmetric under translations and rotations of both the object to be picked and the desired place pose. For example, if the pick object is rotated or translated, then the optimal pick action should also rotate or translate. The same is true for the place pose; if the desired place pose changes, then the place action should also transform accordingly. A recently proposed pick and place framework known as Transporter Net (Zeng, Florence, Tompson, Welker, Chien, Attarian, Armstrong, Krasin, Duong, Sindhwani et al., 2021) captures some of these symmetries, but not all. This paper analytically studies the symmetries present in planar robotic pick and place and proposes a method of incorporating equivariant neural models into Transporter Net in a way that captures all symmetries. The new model, which we call Equivariant Transporter Net, is equivariant to both pick and place symmetries and can immediately generalize pick and place knowledge to different pick and place poses. We evaluate the new model empirically and show that it is much more sample-efficient than the non-symmetric version, resulting in a system that can imitate demonstrated pick and place behavior using very few human demonstrations on a variety of imitation learning tasks. [ABSTRACT FROM AUTHOR]

This paper tackles the task of goal-conditioned dynamic manipulation of deformable objects. This task is highly challenging due to its complex dynamics (introduced by object deformation and high-speed action) and strict task requirements (defined by a precise goal specification). To address these challenges, we present Iterative Residual Policy (IRP), a general learning framework applicable to repeatable tasks with complex dynamics. IRP learns an implicit policy via delta dynamics—instead of modeling the entire dynamical system and inferring actions from that model, IRP learns delta dynamics that predict the effects of delta action on the previously observed trajectory. When combined with adaptive action sampling, the system can quickly optimize its actions online to reach a specified goal. We demonstrate the effectiveness of IRP on two tasks: whipping a rope to hit a target point and swinging a cloth to reach a target pose. Despite being trained only in simulation on a fixed robot setup, IRP is able to efficiently generalize to noisy real-world dynamics, new objects with unseen physical properties, and even different robot hardware embodiments, demonstrating its excellent generalization capability relative to alternative approaches. [ABSTRACT FROM AUTHOR]

Reliable obstacle avoidance, which is essential for safe autonomous robot interaction with the real world, raises various challenges such as difficulties with obstacle perception and latent factor cognition impacting multi-modal obstacle avoidance. In this paper, we propose a Depth visUal Ego-motion Learning (DUEL) model, consisting of a cognitive generation network, a policy decision network and a potential partition network, to learn autonomous obstacle avoidance from expert policies. The DUEL model takes advantage of binocular vision to perceive scene depth. This serves as the input to the cognitive generation network which generates obstacle avoidance policies by maximizing its causal entropy. The policy decision network then optimizes the generation of the policies referring to expert policies. The generated obstacle avoidance policies are simultaneously transferred to the potential partition network to capture the latent factors contained within expert policies and perform multi-modal obstacle avoidance. These three core networks iteratively optimize the multi-modal policies relying on causal entropy and mutual information theorems, which are proven theoretically. Experimental comparisons with state-of-the-art models on 7 metrics demonstrate the effectiveness of the DUEL model. It achieves the best performance with an average ADE (Average Displacement Error) of 0.29 and average FDE (Final Displacement Error) of 0.55 across five different scenarios. Results show that the DUEL model can maintain an average obstacle avoidance success rate of 97% for both simulated and real world scenarios with multiple obstacles, demonstrating its success at capturing latent factors from expert policies. Our source codes are available at https://github.com/ACoTAI/DUEL. [ABSTRACT FROM AUTHOR]

Soft robots have been investigated for various applications due to their inherently superior deformability and flexibility compared to rigid-link robots. However, these robots struggle to perform tasks that require on-demand stiffness, that is, exerting sufficient forces within allowable deflection. In addition, the soft and compliant materials also introduce large deformation and non-negligible nonlinearity, which makes the stiffness analysis and modelling of soft robots fundamentally challenging. This paper proposes a modelling framework to investigate the underlying stiffness and the equivalent compliance properties of soft robots under different configurations. Firstly, a modelling and analysis methodology is described based on Lie theory. Here, we derive two sets (the piecewise constant curvature and Cosserat rod model) of compliance models. Furthermore, the methodology can accommodate the nonlinear responses (e.g., bending angles) resulting from elongation of robots. Using this proposed methodology, the general Cartesian stiffness or compliance matrix can be derived and used for configuration-dependent stiffness analysis. The proposed framework is then instantiated and implemented on fluidic-driven soft continuum robots. The efficacy and modelling accuracy of the proposed methodology are validated using both simulations and experiments. [ABSTRACT FROM AUTHOR]

Automated vehicles (AV) heavily depend on robust perception systems. Current methods for evaluating vision systems focus mainly on frame-by-frame performance. Such evaluation methods appear to be inadequate in assessing the performance of a perception subsystem when used within an AV. In this paper, we present a logic—referred to as Spatio-Temporal Perception Logic (STPL)—which utilizes both spatial and temporal modalities. STPL enables reasoning over perception data using spatial and temporal operators. One major advantage of STPL is that it facilitates basic sanity checks on the functional performance of the perception system, even without ground truth data in some cases. We identify a fragment of STPL which is efficiently monitorable offline in polynomial time. Finally, we present a range of specifications for AV perception systems to highlight the types of requirements that can be expressed and analyzed through offline monitoring with STPL. [ABSTRACT FROM AUTHOR]

This paper presents the energy-optimal trajectories for skid-steer rovers on hard ground, without obstacles. We obtain 29 trajectory structures that are sufficient to describe minimum-energy motion, which are enumerated and described geometrically; 28 of these structures are composed of sequences of circular arcs and straight lines; there is also a special structure called whirls consisting of different circular arcs. Our analysis identifies that the turns in the trajectory structures (aside from whirls) are all circular arcs of a particular turning radius, R′, the turning radius at which the inner wheels of a skid-steer rover are not commanded to turn. This work demonstrates its paramount importance in energy-optimal path planning. There has been a lack of analytical energy-optimal trajectory generation for skid-steer rovers, and we address this problem by a novel approach. The equivalency theorem presented in this work shows that all minimum-energy solutions follow the same path irrespective of velocity constraints that may or may not be imposed. This non-intuitive result stems from the fact that with this model of the system the total energy is fully parameterized by the geometry of the path alone. With this equivalency in mind, one can choose velocity constraints to enforce constant power consumption, thus transforming the energy-optimal problem into an equivalent time-optimal problem. Pontryagin's Minimum Principle can then be used to solve the problem. Accordingly, the extremal paths are obtained and enumerated to find the minimum-energy path. Furthermore, our experimental results by using Husky UGV provide the experimental support for the equivalency theorem. [ABSTRACT FROM AUTHOR]

Determining a globally optimal solution of belief space planning (BSP) in high-dimensional state spaces directly is computationally expensive, as it involves belief propagation and objective function evaluation for each candidate action. However, many problems of interest, such as active SLAM, exhibit structure that can be harnessed to expedite planning. Also, in order to choose an optimal action, an exact value of the objective function is not required as long as the actions can be sorted in the same way. In this paper we leverage these two key aspects and present the topological belief space planning (t-bsp) concept that uses topological signatures to perform this ranking for information-theoretic cost functions, considering only topologies of factor graphs that correspond to future posterior beliefs. In particular, we propose a highly efficient topological signature based on the von Neumann graph entropy that is a function of graph node degrees and supports an incremental update. We analyze it in the context of active pose SLAM and derive error bounds between the proposed topological signature and the original information-theoretic cost function. These bounds are then used to provide performance guarantees for t-bsp with respect to a given solver of the original information-theoretic BSP problem. Realistic and synthetic simulations demonstrate drastic speed-up of the proposed method with respect to the state-of-the-art methods while retaining the ability to select a near-optimal solution. A proof of concept of t-bsp is given in a small-scale real-world active SLAM experiment. [ABSTRACT FROM AUTHOR]

Understanding elastic instability has been a recent focus of concentric tube robot research. Modeling advances have enabled prediction of when instabilities will occur and produced metrics for the stability of the robot during use. In this paper, we show how these metrics can be used to resolve redundancy to avoid elastic instability, opening the door for the practical use of higher curvature designs than have previously been possible. We demonstrate the effectiveness of the approach using a three-tube robot that is stabilized by redundancy resolution when following trajectories that would otherwise result in elastic instabilities. We also show that it is stabilized when teleoperated in ways that otherwise produce elastic instabilities. Lastly, we show that the redundancy resolution framework presented here can be applied to other control objectives useful for surgical robots, such as maximizing or minimizing compliance in desired directions. [ABSTRACT FROM AUTHOR]

In recent years, advancements in Light Detection and Ranging (LiDAR) technology have made 3D LiDAR sensors more compact, lightweight, and affordable. This progress has spurred interest in integrating LiDAR with sensors such as Inertial Measurement Units (IMUs) and cameras for Simultaneous Localization and Mapping (SLAM) research. Public datasets covering different scenarios, platforms, and viewpoints are crucial for multi-sensor fusion SLAM studies, yet most focus on handheld or vehicle-mounted devices with front or 360-degree views. Data from aerial vehicles with downward-looking views is scarce, existing relevant datasets usually feature low altitudes and are mostly limited to small campus environments. To fill this gap, we introduce the Multi-sensor Aerial Robots SLAM dataset (MARS-LVIG dataset), providing unique aerial downward-looking LiDAR-Visual-Inertial-GNSS data with viewpoints from altitudes between 80 m and 130 m. The dataset not only offers new aspects to test and evaluate existing SLAM algorithms, but also brings new challenges which can facilitate researches and developments of more advanced SLAM algorithms. The MARS-LVIG dataset contains 21 sequences, acquired across diversified large-area environments including an aero-model airfield, an island, a rural town, and a valley. Within these sequences, the UAV has speeds varying from 3 m/s to 12 m/s, a scanning area reaching up to 577,000 m2, and the max path length of 7.148 km in a single flight. This dataset encapsulates data collected by a lightweight, hardware-synchronized sensor package that includes a solid-state 3D LiDAR, a global-shutter RGB camera, IMUs, and a raw message receiver of the Global Navigation Satellite System (GNSS). For algorithm evaluation, this dataset releases ground truth of both localization and mapping, which are acquired by on-board Real-time Kinematic (RTK) and DJI L1 (post-processed by its supporting software DJI Terra), respectively. The dataset can be downloaded from: https://mars.hku.hk/dataset.html. [ABSTRACT FROM AUTHOR]

We present a lazy incremental search algorithm, Lifelong-GLS (L-GLS), along with its bounded suboptimal version, Bounded L-GLS (B-LGLS) that combine the search efficiency of incremental search algorithms with the evaluation efficiency of lazy search algorithms for fast replanning in problem domains where edge evaluations are more expensive than vertex expansions. The proposed algorithms generalize Lifelong Planning A* (LPA*) and its bounded suboptimal version, Truncated LPA* (TLPA*), within the Generalized Lazy Search (GLS) framework, so as to restrict expensive edge evaluations only to the current shortest subpath when the cost-to-come inconsistencies are propagated during repair. We also present dynamic versions of the L-GLS and B-LGLS algorithms, called Generalized D* (GD*) and Bounded Generalized D* (B-GD*), respectively, for efficient replanning with non-stationary queries, designed specifically for navigation of mobile robots. We prove that the proposed algorithms are complete and correct in finding a solution that is guaranteed not to exceed the optimal solution cost by a user-chosen factor. Our numerical and experimental results support the claim that the proposed integration of the incremental and lazy search frameworks can help find solutions faster compared to the regular incremental or regular lazy search algorithms when the underlying graph representation changes often. [ABSTRACT FROM AUTHOR]

For real-world applications, autonomous mobile robotic platforms must be capable of navigating safely in a multitude of different and dynamic environments with accurate and robust localization being a key prerequisite. To support further research in this domain, we present the INSANE datasets (Increased Number of Sensors for developing Advanced and Novel Estimators)—a collection of versatile Micro Aerial Vehicle (MAV) datasets for cross-environment localization. The datasets provide various scenarios with multiple stages of difficulty for localization methods. These scenarios range from trajectories in the controlled environment of an indoor motion capture facility, to experiments where the vehicle performs an outdoor maneuver and transitions into a building, requiring changes of sensor modalities, up to purely outdoor flight maneuvers in a challenging Mars analog environment to simulate scenarios which current and future Mars helicopters would need to perform. The presented work aims to provide data that reflects real-world scenarios and sensor effects. The extensive sensor suite includes various sensor categories, including multiple Inertial Measurement Units (IMUs) and cameras. Sensor data is made available as unprocessed measurements and each dataset provides highly accurate ground truth, including the outdoor experiments where a dual Real-Time Kinematic (RTK) Global Navigation Satellite System (GNSS) setup provides sub-degree and centimeter accuracy (1-sigma). The sensor suite also includes a dedicated high-rate IMU to capture all the vibration dynamics of the vehicle during flight to support research on novel machine learning-based sensor signal enhancement methods for improved localization. The datasets and post-processing tools are available at: https://sst.aau.at/cns/datasets/insane-dataset/ [ABSTRACT FROM AUTHOR]

We propose a diffusion approximation method to the continuous-state Markov decision processes that can be utilized to address autonomous navigation and control in unstructured off-road environments. In contrast to most decision-theoretic planning frameworks that assume fully known state transition models, we design a method that eliminates such a strong assumption that is often extremely difficult to engineer in reality. We first take the second-order Taylor expansion of the value function. The Bellman optimality equation is then approximated by a partial differential equation, which only relies on the first and second moments of the transition model. By combining the kernel representation of the value function, we design an efficient policy iteration algorithm whose policy evaluation step can be represented as a linear system of equations characterized by a finite set of supporting states. We first validate the proposed method through extensive simulations in 2 D obstacle avoidance and 2.5 D terrain navigation problems. The results show that the proposed approach leads to a much superior performance over several baselines. We then develop a system that integrates our decision-making framework with onboard perception and conduct real-world experiments in both cluttered indoor and unstructured outdoor environments. The results from the physical systems further demonstrate the applicability of our method in challenging real-world environments. [ABSTRACT FROM AUTHOR]

Ground robots require the crucial capability of traversing unstructured and unprepared terrains and avoiding obstacles to complete tasks in real-world robotics applications such as disaster response. When a robot operates in off-road field environments such as forests, the robot's actual behaviors often do not match its expected or planned behaviors, due to changes in the characteristics of terrains and the robot itself. Therefore, the capability of robot adaptation for consistent behavior generation is essential for maneuverability on unstructured off-road terrains. In order to address the challenge, we propose a novel method of self-reflective terrain-aware adaptation for ground robots to generate consistent controls to navigate over unstructured off-road terrains, which enables robots to more accurately execute the expected behaviors through robot self-reflection while adapting to varying unstructured terrains. To evaluate our method's performance, we conduct extensive experiments using real ground robots with various functionality changes over diverse unstructured off-road terrains. The comprehensive experimental results have shown that our self-reflective terrain-aware adaptation method enables ground robots to generate consistent navigational behaviors and outperforms the compared previous and baseline techniques. [ABSTRACT FROM AUTHOR]

Simultaneous localization and mapping (SLAM) and 3D reconstruction have numerous applications for indoor ground wheeled robots such as floor sweeping and food delivery. To advance research in leveraging semantic information and multi-sensor data to enhance the performances of SLAM and 3D reconstruction in complex indoor scenes, we propose a novel and complex indoor dataset named CID-SIMS, where semantic annotated RGBD images, inertial measurement unit (IMU) measurements, and wheel odometer data are provided from a ground wheeled robot viewpoint. The dataset consists of 22 challenging sequences captured in nine different scenes including office building and apartment environments. Notably, our dataset achieves two significant breakthroughs. Firstly, semantic information and multi-sensor data are provided meanwhile for the first time. Secondly, GeoSLAM is utilized for the first time to generate ground truth trajectories and 3D point clouds within two-centimeter accuracy. With spatial-temporal synchronous ground truth trajectories and 3D point clouds, our dataset is capable of evaluating SLAM and 3D reconstruction algorithms in a unified global coordinate system. We evaluate state-of-the-art SLAM and 3D reconstruction approaches on our dataset, demonstrating that our benchmark is applicable. The dataset is publicly available on https://cid-sims.github.io. [ABSTRACT FROM AUTHOR]

Ultrasound (US) imaging is widely used for biometric measurement and diagnosis of internal organs due to the advantages of being real-time and radiation-free. However, due to inter-operator variations, resulting images highly depend on the experience of sonographers. This work proposes an intelligent robotic sonographer to autonomously "explore" target anatomies and navigate a US probe to standard planes by learning from the expert. The underlying high-level physiological knowledge from experts is inferred by a neural reward function, using a ranked pairwise image comparison approach in a self-supervised fashion. This process can be referred to as understanding the "language of sonography." Considering the generalization capability to overcome inter-patient variations, mutual information is estimated by a network to explicitly disentangle the task-related and domain features in latent space. The robotic localization is carried out in coarse-to-fine mode based on the predicted reward associated with B-mode images. To validate the effectiveness of the proposed reward inference network, representative experiments were performed on vascular phantoms ("line" target), two types of ex vivo animal organ phantoms (chicken heart and lamb kidney representing "point" target), and in vivo human carotids. To further validate the performance of the autonomous acquisition framework, physical robotic acquisitions were performed on three phantoms (vascular, chicken heart, and lamb kidney). The results demonstrated that the proposed advanced framework can robustly work on a variety of seen and unseen phantoms as well as in vivo human carotid data. Code : https://github.com/yuan-12138/MI-GPSR. Video : https://youtu.be/u4ThAA9onE0. [ABSTRACT FROM AUTHOR]

Autonomous navigation and information gathering in challenging environments are demanding since the robot's sensors may be susceptible to non-negligible noise, its localization and mapping may be subject to significant uncertainty and drift, and performing collision-checking or evaluating utility functions using a map often requires high computational costs. We propose a learning-based method to efficiently tackle this problem without relying on a map of the environment or the robot's position. Our method utilizes a Collision Prediction Network (CPN) for predicting the collision scores of a set of action sequences, and an Information gain Prediction Network (IPN) for estimating their associated information gain. Both networks assume access to a) the depth image (CPN) or the depth image and the detection mask from any visual method (IPN), b) the robot's partial state (including its linear velocities, z -axis angular velocity, and roll/pitch angles), and c) a library of action sequences. Specifically, the CPN accounts for the estimation uncertainty of the robot's partial state and the neural network's epistemic uncertainty by using the Unscented Transform and an ensemble of neural networks. The outputs of the networks are combined with a goal vector to identify the next-best-action sequence. Simulation studies demonstrate the method's robustness against noisy robot velocity estimates and depth images, alongside its advantages compared to state-of-the-art methods and baselines in (visually-attentive) navigation tasks. Lastly, multiple real-world experiments are presented, including safe flights at 2.5 m/s in a cluttered corridor, and missions inside a dense forest alongside visually-attentive navigation in industrial and university buildings. [ABSTRACT FROM AUTHOR]

The convergence of many numerical optimization techniques is highly dependent on the initial guess given to the solver. To address this issue, we propose a novel approach that utilizes tensor methods to initialize existing optimization solvers near global optima. Our method does not require access to a database of good solutions. We first transform the cost function, which depends on both task parameters and optimization variables, into a probability density function. Unlike existing approaches, the joint probability distribution of the task parameters and optimization variables is approximated using the Tensor Train model, which enables efficient conditioning and sampling. We treat the task parameters as random variables, and for a given task, we generate samples for decision variables from the conditional distribution to initialize the optimization solver. Our method can produce multiple solutions (when they exist) faster than existing methods. We first evaluate the approach on benchmark functions for numerical optimization that are hard to solve using gradient-based optimization solvers with a naive initialization. The results show that the proposed method can generate samples close to global optima and from multiple modes. We then demonstrate the generality and relevance of our framework to robotics by applying it to inverse kinematics with obstacles and motion planning problems with a 7-DoF manipulator. [ABSTRACT FROM AUTHOR]

Simultaneous localization and mapping (SLAM) has been an active research problem over recent decades. Many leading solutions are available that can achieve remarkable performance in environments with familiar structure, such as indoors and cities. However, our work shows that these leading systems fail in an agricultural setting, particularly in under the canopy navigation in the largest-in-acreage crops of the world: corn (Zea mays) and soybean (Glycine max). The presence of plenty of visual clutter due to leaves, varying illumination, and stark visual similarity makes these environments lose the familiar structure on which SLAM algorithms rely on. To advance SLAM in such unstructured agricultural environments, we present a comprehensive agricultural dataset. Our open dataset consists of stereo images, IMUs, wheel encoders, and GPS measurements continuously recorded from a mobile robot in corn and soybean fields across different growth stages. In addition, we present best-case benchmark results for several leading visual-inertial odometry and SLAM systems. Our data and benchmark clearly show that there is significant research promise in SLAM for agricultural settings. The dataset is available online at: https://github.com/jrcuaranv/terrasentia-dataset. [ABSTRACT FROM AUTHOR]

Designing reward functions is a difficult task in AI and robotics. The complex task of directly specifying all the desirable behaviors a robot needs to optimize often proves challenging for humans. A popular solution is to learn reward functions using expert demonstrations. This approach, however, is fraught with many challenges. Some methods require heavily structured models, for example, reward functions that are linear in some predefined set of features, while others adopt less structured reward functions that may necessitate tremendous amounts of data. Moreover, it is difficult for humans to provide demonstrations on robots with high degrees of freedom, or even quantifying reward values for given trajectories. To address these challenges, we present a preference-based learning approach, where human feedback is in the form of comparisons between trajectories. We do not assume highly constrained structures on the reward function. Instead, we employ a Gaussian process to model the reward function and propose a mathematical formulation to actively fit the model using only human preferences. Our approach enables us to tackle both inflexibility and data-inefficiency problems within a preference-based learning framework. We further analyze our algorithm in comparison to several baselines on reward optimization, where the goal is to find the optimal robot trajectory in a data-efficient way instead of learning the reward function for every possible trajectory. Our results in three different simulation experiments and a user study show our approach can efficiently learn expressive reward functions for robotic tasks, and outperform the baselines in both reward learning and reward optimization. [ABSTRACT FROM AUTHOR]

Muscles in animals and actuation systems in advanced robots consist not of the actuation component alone; the motive, dissipative, and proprioceptive components exist in a complete set to achieve versatile and precise manipulation tasks. We present such a system as a linear electrostatic actuator package incorporated with sensing and braking components. Our modular actuator design is composed of these actuator films and a dielectric fluid, and we examine the performance of the proposed system both theoretically and experimentally. In addition, we introduce a mechanism of optical proprioceptive sensing utilizing the Moiré pattern innately generated on the actuator surface, which allows high-resolution reading of the position of the actuator without noise. The optical sensor is also capable of measuring the force exerted by the actuator. Lastly, we add an electroadhesive brake in the package in parallel with the actuator, introducing a method of mode switching that utilizes all three components and presenting control demonstrations with a robot arm. Our actuation system is compact and flexible and can be easily integrated with various robotic applications. [ABSTRACT FROM AUTHOR]

The agricultural industry is being transformed, thanks to recent innovations in computer vision and deep learning. However, the lack of specific datasets collected in natural agricultural environments is, arguably, the main bottleneck for novel discoveries and benchmarking. The present work provides a novel dataset, Magro, and a framework to expand data collection. We present the first version of the Magro Dataset V1.0, consisting of nine ROS bags (and the corresponding raw data) containing data collected in apple and pear crops. Data were gathered, repeating a fixed trajectory on different days under different illumination and weather conditions. To support the evaluation of loop closure algorithms, the trajectories are designed to have loop closures, revisiting some places from different viewpoints. We use a Clearpath's Jackal robot equipped with stereo cameras pointing to the front and left side, a 3D LIDAR, three inertial measurement units (IMU), and wheel encoders. Additionally, we provide calibrated RTK GPS data that can be used as ground truth. Our dataset is openly available, and it will be updated to have more data and variability. Finally, we tested two existing state-of-the-art algorithms for vision and point cloud-based localization and mapping on our novel dataset to validate the dataset's usability. [ABSTRACT FROM AUTHOR]

Social robots facilitate improved human–robot interactions through nonverbal behaviors such as handshakes or hugs. However, the traditional methods, which rely on precoded motions, are predictable and can detract from the perception of robots as interactive agents. To address this issue, we have introduced a Seq2Seq-based neural network model that learns social behaviors from human–human interactions in an end-to-end manner. To mitigate the risk of invalid pose sequences during long-term behavior generation, we incorporated a generative adversarial network (GAN). This proposed method was tested using the humanoid robot, Pepper, in a simulated environment. Given the challenges in assessing the success of social behavior generation, we devised novel metrics to quantify the discrepancy between the generated and ground-truth behaviors. Our analysis reveals the impact of different networks on behavior generation performance and compares the efficacy of learning multiple behaviors versus a single behavior. We anticipate that our method will find application in various sectors, including home service, guide, delivery, educational, and virtual robots, thereby enhancing user interaction and enjoyment. [ABSTRACT FROM AUTHOR]

Global registration is a fundamental task that estimates the relative pose between two viewpoints of 3D point clouds. However, there are two issues that degrade the performance of global registration in LiDAR SLAM: one is the sparsity issue and the other is degeneracy. The sparsity issue is caused by the sparse characteristics of the 3D point cloud measurements in a mechanically spinning LiDAR sensor. The degeneracy issue sometimes occurs because the outlier-rejection methods reject too many correspondences, leaving less than three inliers. These two issues have become more severe as the pose discrepancy between the two viewpoints of 3D point clouds becomes greater. To tackle these problems, we propose a robust global registration framework, called Quatro++. Extending our previous work that solely focused on the global registration itself, we address the robust global registration in terms of the loop closing in LiDAR SLAM. To this end, ground segmentation is exploited to achieve robust global registration. Through the experiments, we demonstrate that our proposed method shows a higher success rate than the state-of-the-art global registration methods, overcoming the sparsity and degeneracy issues. In addition, we show that ground segmentation significantly helps to increase the success rate for the ground vehicles. Finally, we apply our proposed method to the loop closing module in LiDAR SLAM and confirm that the quality of the loop constraints is improved, showing more precise mapping results. Therefore, the experimental evidence corroborated the suitability of our method as an initial alignment in the loop closing. Our code is available at https://quatro-plusplus.github.io. [ABSTRACT FROM AUTHOR]

Modeling and manipulating elasto-plastic objects are essential capabilities for robots to perform complex industrial and household interaction tasks (e.g., stuffing dumplings, rolling sushi, and making pottery). However, due to the high degrees of freedom of elasto-plastic objects, significant challenges exist in virtually every aspect of the robotic manipulation pipeline, for example, representing the states, modeling the dynamics, and synthesizing the control signals. We propose to tackle these challenges by employing a particle-based representation for elasto-plastic objects in a model-based planning framework. Our system, RoboCraft, only assumes access to raw RGBD visual observations. It transforms the sensory data into particles and learns a particle-based dynamics model using graph neural networks (GNNs) to capture the structure of the underlying system. The learned model can then be coupled with model predictive control (MPC) algorithms to plan the robot's behavior. We show through experiments that with just 10 min of real-world robot interaction data, our robot can learn a dynamics model that can be used to synthesize control signals to deform elasto-plastic objects into various complex target shapes, including shapes that the robot has never encountered before. We perform systematic evaluations in both simulation and the real world to demonstrate the robot's manipulation capabilities. [ABSTRACT FROM AUTHOR]

We study the generalized Procrustes analysis (GPA), as a minimal formulation to the simultaneous localization and mapping (SLAM) problem. We propose Kernel-GPA, a novel global registration technique to solve SLAM in the deformable environment. We propose the concept of deformable transformation which encodes the entangled pose and deformation. We define deformable transformations using a kernel method and show that both the deformable transformations and the environment map can be solved globally in closed-form, up to global scale ambiguities. We solve the scale ambiguities by an optimization formulation that maximizes rigidity. We demonstrate Kernel-GPA using the Gaussian kernel and validate the superiority of Kernel-GPA with various datasets. Code and data are available at https://bitbucket.org/FangBai/deformableprocrustes. [ABSTRACT FROM AUTHOR]

We present a coordinated autonomy pipeline for multi-sensor exploration of confined environments. We simultaneously address four broad challenges that are typically overlooked in prior work: (a) make effective use of both range and vision sensing modalities, (b) perform this exploration across a wide range of environments, (c) be resilient to adverse events, and (d) execute this onboard teams of physical robots. Our solution centers around a behavior tree architecture, which adaptively switches between various behaviors involving coordinated exploration and responding to adverse events. Our exploration strategy exploits the benefits of both visual and range sensors with a generalized frontier-based exploration algorithm and an OpenVDB-based map processing pipeline. Our local planner utilizes a dynamically feasible trajectory library and a GPU-based Euclidean distance transform map to allow fast and safe navigation through both tight doorways and expansive spaces. The autonomy pipeline is evaluated with an extensive set of field experiments, with teams of up to three robots that fly up to 3 m/s and distances exceeding 1 km in confined spaces. We provide a summary of various field experiments and detail resilient behaviors that arose: maneuvering narrow doorways, adapting to unexpected environment changes, and emergency landing. Experiments are also detailed from the DARPA Subterranean Challenge, where our proposed autonomy pipeline contributed to us winning the "Most Sectors Explored" award. We provide an extended discussion of lessons learned, release software as open source, and present a video that illustrates our extensive field trials. [ABSTRACT FROM AUTHOR]

Manipulating volumetric deformable objects in the real world, like plush toys and pizza dough, brings substantial challenges due to infinite shape variations, non-rigid motions, and partial observability. We introduce ACID, an action-conditional visual dynamics model for volumetric deformable objects based on structured implicit neural representations. ACID integrates two new techniques: implicit representations for action-conditional dynamics and geodesics-based contrastive learning. To represent deformable dynamics from partial RGB-D observations, we learn implicit representations of occupancy and flow-based forward dynamics. To accurately identify state change under large non-rigid deformations, we learn a correspondence embedding field through a novel geodesics-based contrastive loss. To evaluate our approach, we develop a simulation framework for manipulating complex deformable shapes in realistic scenes and a benchmark containing over 17,000 action trajectories with six types of plush toys and 78 variants. Our model achieves the best performance in geometry, correspondence, and dynamics predictions over existing approaches. The ACID dynamics models are successfully employed for goal-conditioned deformable manipulation tasks, resulting in a 30% increase in task success rate over the strongest baseline. Furthermore, we apply the simulation-trained ACID model directly to real-world objects and show success in manipulating them into target configurations. https://b0ku1.github.io/acid/ [ABSTRACT FROM AUTHOR]

Robotic Information Gathering (RIG) is a foundational research topic that answers how a robot (team) collects informative data to efficiently build an accurate model of an unknown target function under robot embodiment constraints. RIG has many applications, including but not limited to autonomous exploration and mapping, 3D reconstruction or inspection, search and rescue, and environmental monitoring. A RIG system relies on a probabilistic model's prediction uncertainty to identify critical areas for informative data collection. Gaussian processes (GPs) with stationary kernels have been widely adopted for spatial modeling. However, real-world spatial data is typically non-stationary —different locations do not have the same degree of variability. As a result, the prediction uncertainty does not accurately reveal prediction error, limiting the success of RIG algorithms. We propose a family of non-stationary kernels named Attentive Kernel (AK), which is simple and robust and can extend any existing kernel to a non-stationary one. We evaluate the new kernel in elevation mapping tasks, where AK provides better accuracy and uncertainty quantification over the commonly used stationary kernels and the leading non-stationary kernels. The improved uncertainty quantification guides the downstream informative planner to collect more valuable data around the high-error area, further increasing prediction accuracy. A field experiment demonstrates that the proposed method can guide an Autonomous Surface Vehicle (ASV) to prioritize data collection in locations with significant spatial variations, enabling the model to characterize salient environmental features. [ABSTRACT FROM AUTHOR]

In endoscopic submucosal dissection (ESD), the gastrointestinal (GI) tract warrants the surgical instruments to navigate through a long, narrow and tortuous endoscope. This poses a great challenge in developing ESD instruments with small dimensions, flexibility, and high distal dexterity. In this work, we propose the first Transendoscopic Flexible Parallel Continuum Robotic mechanism to develop a miniature dexterous flexible-stiff-balanced Wrist (FPCW). Besides, it can steer multifunctional instruments of diameters 2.5 mm to 3.5 mm, including the electrosurgical knife, injection needle, and forceps. Our FPCW instruments are adaptable to commercially available dual-channel endoscopes (diameter: <12 mm, channel width: 2.8 mm and around 3.8 mm). Furthermore, we develop a surgical telerobotic system, called DREAMS (Dual-arm Robotic Endoscopic Assistant for Minimally Invasive Surgery), by using our smallest FPCW instruments for bimanual ESD procedures. First, we conduct a series of experiments to determine the FPCW's design and kinematics parameters and to verify the mechanical properties of the FPCW instruments' prototypes, including workspace, stiffness, strength, and teleoperation accuracy. Second, we validate the functionality of the FPCW instruments through ex-vivo tests by performing ESD steps on porcine stomachs. Finally, we perform an invivo test on a live porcine model and showcase that our developed DREAMS can be teleoperated intuitively to perform bimanual ESD efficiently with an average dissection speed of 108.95 mm2/min at the greater curvature in gastric body, which demonstrates that our DREAMS has satisfactory maneuverability as well as accuracy and is more competitive than counterpart robotic systems. [ABSTRACT FROM AUTHOR]

We address the theoretical and practical problems related to the trajectory generation and tracking control of tail-sitter UAVs. Theoretically, we focus on the differential flatness property with full exploitation of actual UAV aerodynamic models, which lays a foundation for generating dynamically feasible trajectory and achieving high-performance tracking control. We have found that a tail-sitter is differentially flat with accurate (not simplified) aerodynamic models within the entire flight envelope, by specifying coordinate flight condition and choosing the vehicle position as the flat output. This fundamental property allows us to fully exploit the high-fidelity aerodynamic models in the trajectory planning and tracking control to achieve accurate tail-sitter flights. Particularly, an optimization-based trajectory planner for tail-sitters is proposed to design high-quality, smooth trajectories with consideration of kinodynamic constraints, singularity-free constraints, and actuator saturation. The planned trajectory of flat output is transformed into state trajectory in real time with optional consideration of wind in environments. To track the state trajectory, a global, singularity-free, and minimally parameterized on-manifold MPC is developed, which fully leverages the accurate aerodynamic model to achieve high-accuracy trajectory tracking within the whole flight envelope. The proposed algorithms are implemented on our quadrotor tail-sitter prototype, "Hong Hu," and their effectiveness are demonstrated through extensive real-world experiments in both indoor and outdoor field tests, including agile SE(3) flight through consecutive narrow windows requiring specific attitude and with speed up to 10 m/s, typical tail-sitter maneuvers (transition, level flight, and loiter) with speed up to 20 m/s, and extremely aggressive aerobatic maneuvers (Wingover, Loop, Vertical Eight, and Cuban Eight) with acceleration up to 2.5 g. The video demonstration is available at https://youtu.be/2x_bLbVuyrk. [ABSTRACT FROM AUTHOR]

Multi-step forceful manipulation tasks, such as opening a push-and-twist childproof bottle, require a robot to make various planning choices that are substantially impacted by the requirement to exert force during the task. The robot must reason over discrete and continuous choices relating to the sequence of actions, such as whether to pick up an object, and the parameters of each of those actions, such how to grasp the object. To enable planning and executing forceful manipulation, we augment an existing task and motion planner with constraints that explicitly consider torque and frictional limits, captured through the proposed forceful kinematic chain constraint. In three domains, opening a childproof bottle, twisting a nut and cutting a vegetable, we demonstrate how the system selects from among a combinatorial set of strategies. We also show how cost-sensitive planning can be used to find strategies and parameters that are robust to uncertainty in the physical parameters. [ABSTRACT FROM AUTHOR]

The learning-from-observation (LfO) paradigm allows a robot to learn how to perform actions by observing human actions. Previous research in top-down learning-from-observation has mainly focused on the industrial domain, which consists only of the real physical constraints between a manipulated tool and the robot's working environment. To extend this paradigm to the household domain, which consists of imaginary constraints derived from human common sense, we introduce the idea of semantic constraints, which are represented similarly to the physical constraints by defining an imaginary contact with an imaginary environment. By studying the transitions between contact states under physical and semantic constraints, we derive a necessary and sufficient set of task representations that provides the upper bound of the possible task set. We then apply the task representations to analyze various actions in top-rated household YouTube videos and real home cooking recordings, classify frequently occurring constraint patterns into physical, semantic, and multi-step task groups, and determine a subset that covers standard household actions. Finally, we design and implement task models, corresponding to these task representations in the subset, with the necessary daemon functions to collect the necessary parameters to perform the corresponding household actions. Our results provide promising directions for incorporating common sense into the robot teaching literature. [ABSTRACT FROM AUTHOR]

We propose a structured prediction approach for robot imitation learning from demonstrations. Among various tools for robot imitation learning, supervised learning has been observed to have a prominent role. Structured prediction is a form of supervised learning that enables learning models to operate on output spaces with complex structures. Through the lens of structured prediction, we show how robots can learn to imitate trajectories belonging to not only Euclidean spaces but also Riemannian manifolds. Exploiting ideas from information theory, we propose a class of loss functions based on the f -divergence to measure the information loss between the demonstrated and reproduced probabilistic trajectories. Different types of f -divergence will result in different policies, which we call imitation modes. Furthermore, our approach enables the incorporation of spatial and temporal trajectory modulation, which is necessary for robots to be adaptive to the change in working conditions. We benchmark our algorithm against state-of-the-art methods in terms of trajectory reproduction and adaptation. The quantitative evaluation shows that our approach outperforms other algorithms regarding both accuracy and efficiency. We also report real-world experimental results on learning manifold trajectories in a polishing task with a KUKA LWR robot arm, illustrating the effectiveness of our algorithmic framework. [ABSTRACT FROM AUTHOR]

We propose a dataset to study the influence of object-specific characteristics on human pick-and-place movements and compare the quality of the motion kinematics extracted by various sensors. This dataset is also suitable for promoting a broader discussion on general learning problems in the hand-object interaction domain, such as intention recognition or motion generation with applications in the Robotics field. The dataset consists of the recordings of 15 subjects performing 80 repetitions of a pick-and-place action under various experimental conditions, for a total of 1200 pick-and-places. The data has been collected thanks to a multimodal setup composed of multiple cameras, observing the actions from different perspectives, a motion capture system, and a wrist-worn inertial measurement unit. All the objects manipulated in the experiments are identical in shape, size, and appearance but differ in weight and liquid filling, which influences the carefulness required for their handling. [ABSTRACT FROM AUTHOR]

Consider deploying a team of robots in order to visit sites in a risky environment (i.e., where a robot might be lost during a traversal), subject to team-based operational constraints such as limits on team composition, traffic throughputs, and launch constraints. We formalize this problem using a graph to represent the environment, enforcing probabilistic survival constraints for each robot, and using a matroid (which generalizes linear independence to sets) to capture the team-based operational constraints. The resulting "Matroid Team Surviving Orienteers" (MTSO) problem has broad applications for robotics such as informative path planning, resource delivery, and search and rescue. We demonstrate that the objective for the MTSO problem has submodular structure, which leads us to develop two polynomial time algorithms which are guaranteed to find a solution with value within a constant factor of the optimum. The second of our algorithms is an extension of the accelerated continuous greedy algorithm, and can be applied to much broader classes of constraints while maintaining bounds on suboptimality. In addition to in-depth analysis, we demonstrate the efficiency of our approaches by applying them to a scenario where a team of robots must gather information while avoiding dangers in the Coral Triangle and characterize scaling and parameter selection using a synthetic dataset. [ABSTRACT FROM AUTHOR]

High-dimensional motion planning problems can often be solved significantly faster by using multilevel abstractions. While there are various ways to formally capture multilevel abstractions, we formulate them in terms of fiber bundles. Fiber bundles essentially describe lower-dimensional projections of the state space using local product spaces, which allows us to concisely describe and derive novel algorithms in terms of bundle restrictions and bundle sections. Given such a structure and a corresponding admissible constraint function, we develop highly efficient and asymptotically optimal sampling-based motion planning methods for high-dimensional state spaces. Those methods exploit the structure of fiber bundles through the use of bundle primitives. Those primitives are used to create novel bundle planners, the rapidly-exploring quotient space trees (QRRT*), and the quotient space roadmap planner (QMP*). Both planners are shown to be probabilistically complete and almost-surely asymptotically optimal. To evaluate our bundle planners, we compare them against classical sampling-based planners on benchmarks of four low-dimensional scenarios, and eight high-dimensional scenarios, ranging from 21 to 100 degrees of freedom, including multiple robots and nonholonomic constraints. Our findings show improvements up to two to six orders of magnitude and underline the efficiency of multilevel motion planners and the benefit of exploiting multilevel abstractions using the terminology of fiber bundles. [ABSTRACT FROM AUTHOR]