Search

Controlling biomimetic underwater robots in unknown flow fields remains a challenge due to the strong nonlinearity of the fluid. This article investigates the attitude holding task of a robotic fish swimming in reality. Such a typical sensing-based control task requires the fish to keep a desired angle of attack in an unknown and even varied incoming flow. To this end, we propose a learning-based approach by using a deep neural network directly maps the raw data of sensors equipped on the robot to the continuous control signals in an end-to-end manner. First, based on experimental data of the physical robot, we construct a data-driven simulation environment including three modules of dynamic, sensor, and control. The dynamic and sensor modules are established to model the dynamics of the fish and to generate its sensors’ data, based on which a deep reinforcement learning (DRL) algorithm in the control module is trained to get a control policy. Then, we directly deploy the trained policy to a physical robotic fish for attitude holding task. Experimental results demonstrate the robustness and effectiveness of the DRL policy and, thus, verify the success of our approach to achieving sim-to-real transfer. [ABSTRACT FROM AUTHOR]

This article introduces the analysis and control of a meso-scale two degree-of-freedom robotic endoscopic tool body for minimally invasive surgeries. The design of the robotic tool uses two types of a tendon-driven joint known as a bending flexure joint that allows us to control each degree-of-freedom by minimizing interjoint coupling by design. Pure kinematic modeling and control for these robots may not provide precise control performance due to kinematic uncertainties arising from tendon elongation, tendon slacking, gear backlash, etc. We propose a static model for each of the joints of the robotic tool that avoids several of these problems. Depending on the direction of tendon tension application, the proximal joint displays considerable hysteresis due to the superelastic material characteristics and this is included in our static model. The statics of a highly compliant distal joint is also modeled and validated using finite element analysis and experimental data. Using these models, we develop a control system that comprises of a disturbance observer and the proposed static model to provide precise force control and compensate for joint hysteresis. [ABSTRACT FROM AUTHOR]

In this article, we consider the problem of mass transport cloaking using mobile robots. The robots move along a predefined curve that encloses a safe zone and carry sources that collectively counteract a chemical agent released in the environment. The goal is to steer the mass flux around a desired region so that it remains unaffected by the external concentration. We formulate the problem of controlling the robot positions and release rates as a partial differential equation (PDE)-constrained optimization, where the propagation of the chemical is modeled by the advection-diffusion (AD) PDE. We use a neural network (NN) to approximate the solution of the PDE. Particularly, we propose a novel loss function for the NN that utilizes the variational form of the AD-PDE and allows us to reformulate the planning problem as an unsupervised model-based learning problem. Our loss function is discretization-free and highly parallelizable. Unlike passive cloaking methods that use metamaterials to steer the mass flux, our method is the first to use mobile robots to actively control the concentration levels and create safe zones independent of environmental conditions. We demonstrate the performance of our method in simulations. [ABSTRACT FROM AUTHOR]

This paper presents a method for designing the gait of a snake robot that moves in a complicated environment. We propose a method for expressing the target form of a snake robot by connecting curve segments whose curvature and torsion are already known. Because the characteristics of each combined shape are clear, we can design the target form intuitively and approximate a snake robot configuration to this form with low computational cost. In addition, we propose two novel gaits for the snake robot as a design example of the proposed method. The first gait is aimed at moving over a flange on a pipe, while the other is the crawler gait aimed at moving over rough terrain. We demonstrated the effectiveness of the two gaits on a pipe and rough terrain in experiments. [ABSTRACT FROM AUTHOR]

We propose a framework to generate static robot configurations satisfying a set of physical and geometrical constraints. This is done by formulating nonlinear constrained optimization problems over non-Euclidean manifolds and solving them. To do so, we present a new sequential quadratic programming solver working natively on general manifolds and propose an interface to easily formulate the problems, with the tedious and error-prone work automated for the user. We also introduce several new types of constraints for having more complex contacts or working on forces/torques. Our approach allows an elegant mathematical description of the constraints and we exemplify it through formulation and computation examples in complex scenarios with humanoid robots. [ABSTRACT FROM AUTHOR]

Grasping objects that are too large to envelop is traditionally achieved using friction that is activated by squeezing. We present a family of shear-activated grippers that can grasp such objects without the need to squeeze. When a shear force is applied to the gecko-inspired material in our grippers, adhesion is turned on; this adhesion in turn results in adhesion-controlled friction, a friction force that depends on adhesion rather than a squeezing normal force. Removal of the shear force eliminates adhesion, allowing easy release of an object. A compliant shear-activated gripper without active sensing and control can use the same light touch to lift objects that are soft, brittle, fragile, light, or very heavy. We present three grippers, the first two designed for curved objects, and the third for nearly any shape. Simple models describe the grasping process, and empirical results verify the models. The grippers are demonstrated on objects with a variety of shapes, materials, sizes, and weights. [ABSTRACT FROM AUTHOR]

Robotics have long sought an actuation technology comparable to or as capable as biological muscle tissue. Natural muscles exhibit a high power-to-weight ratio, inherent compliance and damping, fast action, and a high dynamic range. They also produce joint displacements and forces without the need for gearing or additional hardware. Recently, supercoiled commercially available polymer threads (sewing thread or nylon fishing lines) have been used to create significant mechanical power in a muscle-like form factor. Heating and cooling the polymer threads causes contraction and expansion, which can be utilized for actuation. In this paper, we describe the working principle of supercoiled polymer (SCP) actuation and explore the controllability and properties of these threads. We show that under appropriate environmental conditions, the threads are suitable as a building block for a controllable artificial muscle. We leverage off-the-shelf silver-coated threads to enable rapid electrical heating while the low thermal mass allows for rapid cooling. We utilize both thermal and thermomechanical models for feed-forward and feedback control. The resulting SCP actuator regulates to desired force levels in as little as 28 ms. Together with its inherent stiffness and damping, this is sufficient for a position controller to execute large step movements in under 100 ms. This controllability, high performance, the mechanical properties, and the extremely low material cost are indicative of a viable artificial muscle. [ABSTRACT FROM PUBLISHER]

A small but interesting result of Brockett is extended to the Euclidean group $SE(3)$ and is illustrated by several examples. The result concerns the explicit solution of an optimal control problem on Lie groups, where the control belongs to a Lie triple system in the Lie algebra. The extension allows for an objective function based on an indefinite quadratic form. Applying the result requires explicit knowledge of the Lie triple systems of the Lie algebra $se(3)$. Hence, a complete classification of the Lie triple systems of this Lie algebra is derived. Examples are considered for optimal trajectories in three cases. The first case concerns cars moving in the plane. The second looks at motions that rigidly follow the Bishop frame to a space curve. The final example does not have a particular name as it does not seem to have been studied before. The appendix gives a brief introduction to Screw theory. This is essentially the study of the Lie algebra $se(3)$. [ABSTRACT FROM PUBLISHER]

Reinforcement learning (RL) can be a tool for designing policies and controllers for robotic systems. However, the cost of real-world samples remains prohibitive as many RL algorithms require a large number of samples before learning useful policies. Simulators are one way to decrease the number of required real-world samples, but imperfect models make deciding when and how to trust samples from a simulator difficult. We present a framework for efficient RL in a scenario where multiple simulators of a target task are available, each with varying levels of fidelity. The framework is designed to limit the number of samples used in each successively higher-fidelity/cost simulator by allowing a learning agent to choose to run trajectories at the lowest level simulator that will still provide it with useful information. Theoretical proofs of the framework's sample complexity are given and empirical results are demonstrated on a remote-controlled car with multiple simulators. The approach enables RL algorithms to find near-optimal policies in a physical robot domain with fewer expensive real-world samples than previous transfer approaches or learning without simulators. [ABSTRACT FROM PUBLISHER]

This paper proposes a framework called nonlinear robust internal-loop compensator that enables us to bring nonlinear \mathcal H_\infty optimality to robot controllers in a unified and simple way. Using the framework, a controller designed for the nominal plant can achieve additional robustness by simply adding PID-type auxiliary input to the original control law. Robust performance is guaranteed by the nonlinear \mathcal H_\infty optimality and robust stability is guaranteed by proving the extended disturbance input-to-state stability. Moreover, the framework preserves the passivity property of the original controller. Finally, the performance bound can be predicted and leads to the gain tuning rules. By virtue of the tuning rules, the performance can be tuned using only a single variable. The proposed method was validated through the simulations and experiments. [ABSTRACT FROM PUBLISHER]

We propose local magnetic actuation (LMA) as an approach to robotic actuation for surgical instruments. An LMA actuation unit consists of a pair of diametrically magnetized single-dipole cylindrical magnets, working as magnetic gears across the abdominal wall. In this study, we developed a dynamic model for an LMA actuation unit by extending the theory proposed for coaxial magnetic gears. The dynamic model was used for closed-loop control, and two alternative strategies—using either the angular velocity at the motor or at the load as feedback parameter—were compared. The amount of mechanical power that can be transferred across the abdominal wall at different intermagnetic distances was also investigated. The proposed dynamic model presented a relative error below 7.5% in estimating the load torque from the system parameters. Both the strategies proposed for closed-loop control were effective in regulating the load speed with a relative error below 2% of the desired steady-state value. However, the load-side closed-loop control approach was more precise and allowed the system to transmit larger values of torque, showing, at the same time, less dependence from the angular velocity. In particular, an average value of 1.5 mN$\cdot$m can be transferred at 7 cm, increasing up to 13.5 mN$\cdot$m as the separation distance is reduced down to 2 cm. Given the constraints in diameter and volume for a surgical instrument, the proposed approach allows for transferring a larger amount of mechanical power than what would be possible to achieve by embedding commercial dc motors. [ABSTRACT FROM PUBLISHER]

Intuitive and effective physical assistance is an essential requirement for robots sharing their workspace with humans. Application domains reach from manufacturing and service robotics via rehabilitation and mobility aids to education and training. In this context, assistance based on human behavior anticipation has shown superior performance in terms of human effort minimization. However, when a robot's expectations mismatch a human intentions, undesired interaction forces appear incurring safety risks and discomfort. Human behavior prediction is, therefore, a crucial issue: It enables effective anticipation but potentially produces disagreements when prediction errors occur. In this paper, we present a novel control scheme for anticipatory haptic assistance where robot behavior adapts to prediction uncertainty. Following a data-driven stochastic modeling approach, robot assistance is synthesized solving a risk-sensitive optimal control problem, where the cost function and plant dynamics are affected by model uncertainty. The proposed approach is objectively and subjectively evaluated in an experiment with human users. Results indicate that our method outperforms other assistive control approaches in terms of perceived helpfulness and human effort minimization. [ABSTRACT FROM PUBLISHER]

Helical microswimmers that are capable of propulsion at low Reynolds numbers have great potential for numerous applications. Several kinds of artificial magnetic-actuated helical microswimmers have been designed by researchers. However, they are primarily open-loop controlled. This paper aims to investigate methods of closed-loop control of a magnetic-actuated helical swimmer at low Reynolds number by using visual feedback. For many in-vitro applications, helical swimmers should pass through a defined path, for example along channels with no prerequisite on the velocity profile along the path. Therefore, the main objective of this paper is to achieve a velocity-independent planar path following task. Since the planar path following is based on 3-D steering control of the helical swimmer, a 3-D pose estimation of a helical swimmer is introduced based on the real-time visual tracking with a stereo vision system. The contribution of this paper is in two parts: The 3-D steering of a helical swimmer is demonstrated by visual servo control; and the path following of a straight line with visual servo control is achieved, then compared with open-loop control. We further expect that with this visual servo control method, the helical swimmers will be able to follow reference paths at the microscale. [ABSTRACT FROM PUBLISHER]

This paper presents “1STAR,” which is the first robot that is driven by a single actuator but can be directly commanded to move straight or turn clockwise or counterclockwise. The legged robot relies on a novel actuation gait, which exploits the compliance disparity between alternate stance tripods, to generate rotation by continuously accelerating and decelerating the legs. The direction of turning depends on the configuration of the legs—tripod left or right—and the timing of the acceleration and deceleration. Alternating leg acceleration in successive steps allows for continuous rotation in the desired direction. The turning radius can be varied by changing the timing of the leg acceleration and deceleration without changing the cycle frequency and linear speed. A simplified kinematic motion model of a robot is presented, and a dynamic simulation is performed to analyze the behavior and optimize robot parameters. The locomotion gait is verified experimentally using our newly designed “1STAR” robot. [ABSTRACT FROM PUBLISHER]

This paper proposes an interval-based method for the validation of reliable and robust navigation rules for mobile robots. The main idea is to show that for all feasible perturbations, there exists a safe subset of the state space such that the system cannot escape. The methodology is illustrated on the line-following problem of a sailboat and then validated on an actual experiment where an actual sailboat robot, which is named Vaimos, sails autonomously from Brest to Douarnenez (i.e., more than 100 km). [ABSTRACT FROM AUTHOR]

We present a computational framework for automatic synthesis of control and communication strategies for a robotic team from task specifications that are given as regular expressions about servicing requests in an environment. We assume that the location of the requests in the environment and the robot capacities and cooperation requirements to service the requests are known. Our approach is based on two main ideas. First, we extend recent results from formal synthesis of distributed systems to check for the distributability of the task specification and to generate local specifications, while accounting for the service and communication capabilities of the robots. Second, by using a technique that is inspired by linear temporal logic model checking, we generate individual control and communication strategies. We illustrate the method with experimental results in our robotic urban-like environment. [ABSTRACT FROM PUBLISHER]

The control issue of underwater robots is very challenging due to the nonlinearity, time variance, unpredictable external disturbances, such as the sea current fluctuation, and the difficulty in accurately modeling the hydrodynamic effect. Conventional linear controllers may fail in satisfying performance requirements, especially when changes in the system and environment occur during the operation since it is almost impossible to manually retune the control parameters in water. Therefore, it is highly desirable to have an underwater robot controller capable of self-adjusting control parameters when the overall performance degrades. This paper presents the theory and experimental work of the adaptive plus disturbance observer (ADOB) controller for underwater robots, which is robust with respect to external disturbance and uncertainties in the system. This control scheme consists of disturbance observer (DOB) as the inner-loop controller and a nonregressor based adaptive controller as the outer-loop con troller. The effectiveness of the ADOB was experimentally investigated by implementing three controllers: PID, PID plus DOB, and ADOB on an autonomous underwater robot, ODIN III. [ABSTRACT FROM AUTHOR]