Search

Rapid urbanization has led to resource shortages, necessitating sustainable approaches in the building industry. This research proposes a preliminary voxel-based modular-architectural design strategy (VMADS), focusing on reusable “H” blocks for component connections and construction. By integrating computational design and robotic fabrication, VMADS enhances precision and efficiency. The framework addresses discrete building theory, prefabrication, and autonomous architecture, emphasizing wood’s anisotropic nature for structural integrity. Experimental results validated VMADS through digital simulations and physical tests, demonstrating its potential to create sustainable, reconfigurable structures and revolutionize construction practices.

In modern industrial production, robotic assembly systems play a crucial role. As robots take on more tasks, the need for formal methods arises to define, control, and execute these tasks. This paper introduces a comprehensive approach to designing and generating control code for robotic assembly systems, taking task sequence planning into account. This methodology utilizes Petri nets (PNs) as a formal modeling and synthesis tool for the controller. Initially, the task sequences for assembly operations are represented using PN formalism. Supervisors are then synthesized for task sequence control specifications. Finally, the control code is obtained by the proposed methodology for industrial robots. By implementing this supervisory control structure, real-time control of the robotic assembly system is achieved. Experimental studies were conducted using an assembly cell equipped with an industrial robot. This methodology bridges the gap between the design and implementation of formal controllers for industrial robots. The proposed approach integrates formal methods into robot programming to leverage several advantages, including correctness assurance, complexity handling, improved documentation and clarity, enhanced safety and reliability, property verification, and scalability.

In robotic assembly of flexible flat cables (FFCs), a unique challenge is the inherent difficulty in manipulating such flexible objects compared to their rigid counterparts and the precise estimation of the cable pose. This work proposes a framework that combines object pose estimation using computer-aided design (CAD) models and multiview fusion to perform precise FFC assembly. Our key insight is that a multiview fusion combined with pretrained 6-D pose estimation models offers a more flexible and precise object pose estimation. In a series of experiments involving FFC insertion tasks requiring assembly tolerances down to 0.1 mm, our approach achieves an insertion success rate of 399 out of 400 total attempts. Furthermore, the assembly tasks include the releasing and securing of FFCs from cable connectors, where the system is successful in 200 out of 200 trials. We have also demonstrated the generalization capability of the methodology by successfully completing insertion tasks for common electronic cables like DisplayPort and USB-A, achieving 199 successes in 200 trials. The results not only validate the feasibility of the proposed approach, but also demonstrate its robustness for real-world industrial applications.

IntroductionRobotic assembly tasks require precise manipulation and coordination, often necessitating advanced learning techniques to achieve efficient and effective performance. While residual reinforcement learning with a base policy has shown promise in this domain, existing base policy approaches often rely on hand-designed full-state features and policies or extensive demonstrations, limiting their applicability in semi-structured environments.MethodsIn this study, we propose an innovative Object-Embodiment-Centric Imitation and Residual Reinforcement Learning (OEC-IRRL) approach that leverages an object-embodiment-centric (OEC) task representation to integrate vision models with imitation and residual learning. By utilizing a single demonstration and minimizing interactions with the environment, our method aims to enhance learning efficiency and effectiveness. The proposed method involves three key steps: creating an object-embodiment-centric task representation, employing imitation learning for a base policy using via-point movement primitives for generalization to different settings, and utilizing residual RL for uncertainty-aware policy refinement during the assembly phase.ResultsThrough a series of comprehensive experiments, we investigate the impact of the OEC task representation on base and residual policy learning and demonstrate the effectiveness of the method in semi-structured environments. Our results indicate that the approach, requiring only a single demonstration and less than 1.2 h of interaction, improves success rates by 46% and reduces assembly time by 25%.DiscussionThis research presents a promising avenue for robotic assembly tasks, providing a viable solution without the need for specialized expertise or custom fixtures.

Human–robot (HR) collaboration (HRC) is an emerging research field because of the complementary advantages of humans and robots. An HRC framework for robotic assembly based on impedance control is proposed in this paper. In the HRC framework, the human is the decision maker, the robot acts as the executor, while the assembly environment provides constraints. The robot is the main executor to perform the assembly action, which has the position control, drag and drop, positive impedance control, and negative impedance control modes. To reveal the characteristics of the HRC framework, the switch condition map of different control modes and the stability analysis of the HR coupled system are discussed. In the end, HRC assembly experiments are conducted, where the HRC assembly task can be accomplished when the assembling tolerance is 0.08 mm or with the interference fit. Experiments show that the HRC assembly has the complementary advantages of humans and robots and is efficient in finishing complex assembly tasks.

Robotic systems are crucial in modern manufacturing. Complex assembly tasks require the collaboration of multiple robots. Their orchestration is challenging due to tight tolerances and precision requirements. In this work, we set up two Franka Panda robots performing a peg-in-hole insertion task of 1 mm clearance. We structure the control system hierarchically, planning the robots’ feedback-based trajectories with a central policy trained with reinforcement learning. These trajectories are executed by a low-level impedance controller on each robot. To enhance training convergence, we use reverse curriculum learning, novel for such a two-armed control task, iteratively structured with a minimum requirements and fine-tuning phase. We incorporate domain randomization, varying initial joint configurations of the task for generalization of the applicability. After training, we test the system in a simulation, discovering the impact of curriculum parameters on the emerging process time and its variance. Finally, we transfer the trained model to the real-world, resulting in a small decrease in task duration. Comparing our approach to classical path planning and control shows a decrease in process time, but higher robustness towards calibration errors.

The visual measurement of deep holes in composite material workpieces constitutes a critical step in the robotic assembly of aerospace components. The positioning accuracy of assembly holes significantly impacts the assembly quality of components. However, the complex texture of the composite material surface and mutual interference between the imaging of the inlet and outlet edges of deep holes significantly challenge hole detection. A visual measurement method for deep holes in composite materials based on the radial penalty Laplacian operator is proposed to address the issues by suppressing visual noise and enhancing the features of hole edges. Coupled with a novel inflection-point-removal algorithm, this approach enables the accurate detection of holes with a diameter of 10 mm and a depth of 50 mm in composite material components, achieving a measurement precision of 0.03 mm.

To enhance the accuracy of robotic assembly planning by understanding the graphical instruction manual, this paper proposes a novel two-step error correction method. While constructing the Assembly Task Sequence Graph (ATSG) from the instruction manual, we performed an error correction focusing on the component, symbol, speech bubble, and model number included in the manual. The component and symbol information were used to check the correctness of the manipulated components, the needed motion, and tool used in a single step of the assembly task. The speech bubble information was used to remove the repeatedly drawn components on the instruction images. The model number was recognized to distinguish components that have different models but were classified into the same class in the object detection procedure. After constructing the ATSG, we additionally performed the error correction by checking the total number of components used for the assembly task. The effectiveness of the proposed method was verified by comparing the ATSG generated from four different error correction methods of five different chairs. The results show that the proposed method decreased the influence of wrong detection results to generate an ATSG with higher accuracy and generalized the previous ATSG for different types of chairs. Finally, we built an extended ATSG from the graphical instruction manual of a kid chair, and the chair assembly task was performed by a dual-arm robot.

To achieve precise and time-efficient operation to ensure stable and reliable product assembly, custom-made jigs have been developed for each assembly part. However, designing these jigs is a time-consuming task. Therefore, this study proposes a jamming gripper-inspired flexible jig using a soft membrane consisting of transparent beads and oil with an adjusted refractive index. The proposed jig is equipped with a hydraulic drive system with visual control of the amount of transparent oil inside the jig. The proposed system enables parts fixing by creating a jammed state while maintaining optical transparency, thereby facilitating the visual sensing of the jig’s membrane from camera sensors embedded in the jig. Furthermore, a sensing method was developed to perform a precise assembly without external sensors to estimate the pose of an object based on the proposed point-to-function iterative closest point. The results show that the proposed system successfully estimates the orientation at a root mean square error of less than ang 4 when the position is varied from $\mathrm {0 mm}$ to $\mathrm {20 mm}$ and the angle is varied from ang 0 to ang 20. Furthermore, it can repeatedly fix the object in any orientation from ang 0 to ang 20 at a position of less than $\mathrm {0.5 mm}$ and an angle of less than ang 1.1.

Aeronautical robotic applications use quite large, heavy robots with huge end effectors that are frequently multifunctional. An assembly jig to hold a fuselage panel and two medium-sized six-axis robots fixed on linear axes, referred to as the internal and the external robot with respect to the curvature of the panel, make up the Lean robotized AssemBly and cOntrol of composite aeRostructures (LABOR) work cell. A distributed software architecture is proposed in which individual modules are developed to execute specific subprocesses, each implementing innovative algorithms that solve the main drawbacks of state-of-the-art solutions. Real-time referencing adopts a point-cloud-based strategy to reconstruct and process the part before drilling, avoiding hole positioning errors. Accurate concentric countersink diameters are made possible through the automatic adjustment of the drilling tool with respect to the skin panel, which guarantees its orthogonality, as well as the implementation of process parameter optimization algorithms based on historical results that compensate for the wear of the drilling bits. Automatic sealing and fastening strategies that involve the measurement of the main fastener quality parameters allow for the complete verification of the entire assembly process of each part. Additionally, an advanced multimodal perception system continuously monitors the collaborative workspace to ensure safe human–robot collaboration (HRC) tasks. Through this integrated architecture, LABOR substantially reduces expenses and facilitates maintenance and programming.

In modern production, automation of assembly processes is achieved by using industrial robots with high positioning accuracy, as well as with elements of tactile adaptation and expensive software. The article proposes an alternative approach to solving the problem of reducing the cost of assembly processes. This approach consists of using production tooling in the form of a soft adaptive assembly module that is mounted on the robot arm. The proposed assembly module allows not only for compensating for errors in the relative orientation of assembly objects, but also excludes the possibility of damage to fragile assembly components or for assembly objects with a thin protective coating of a chemical or galvanic nature. This effect is achieved due to the use of elastic corrugated chambers in the module structure, which, when compressed air is supplied to them, correct the position of one of the assembly objects. The article offers a description of the new design of the assembly module, as well as offers analytical dependencies for calculating the design parameters of the module. The results of modeling the parameters of the assembly module are presented. The main economic effect is to reduce the cost of technological equipment for industrial robots while maintaining the ability to adapt to the conditions of the assembly process.

At present, robotization of assembly processes is achieved through the use of industrial robots with high positioning accuracy in conjunction with tactile means of adaptation to the conditions of assembly of precision parts. The cost of such robots is many times higher than the cost of simple robots with low positioning accuracy of the robot arm. The research in this article is aimed at reducing the cost of assembly processes for precision parts by applying the position correction of the connected parts not by the robot hand, but by an additional technological module that is installed on the manipulator of a simple robot and performs high-speed stochastic mismatch scan of assembly objects. The article presents the results of a full factorial experiment of the process of joining precision cylindrical parts with a gap of no more than 3...5 microns. A regression model of this process is proposed, a formula for calculating the quasi-optimal modes of precision assembly and graphanalytical dependences of the assembly time on the scanning modes of the misalignment of assembly objects are given. The proposed high-speed method for compensating for the positioning error of an industrial robot makes it possible to assemble precision parts in a very short time within 1...3(s). The main economic effect of the research results is that the device for scanning the misalignment of assembly objects, which is installed on the arm of an inexpensive robot with a low positioning accuracy, can significantly increase the assembly speed and reduce capital investments in robotic assembly of high-precision parts.

Complex assembly tasks involve nonlinear and low-clearance insertion trajectories with varying contact forces at different stages. For a robot to solve these tasks, it requires a precise and adaptive controller which conventional force control methods cannot provide. Imitation learning is a promising method for learning controllers that can solve the nonlinear trajectories from human demonstrations without needing to explicitly program them into the robot. However, the force profiles obtain from human demonstration via tele-operation tend to be sub-optimal for complex assembly tasks, thus it is undesirable to imitate such force profiles. Reinforcement learning learns adaptive control policies through interactions with the environment but struggles with low sample efficiency and equipment tear and wear in the physical world. To address these problems, we present a combined learning-based framework to solve complex robotic assembly tasks from human demonstrations via hybrid trajectory learning and force learning. The main contribution of this work is the development of a framework that combines imitation learning, to learn the nominal motion trajectory, with a reinforcement learning-based force control scheme to learn an optimal force control policy, which can satisfy the nominal trajectory while adapting to the force requirement of the assembly task. To further improve the imitation learning part, we develop a hierarchical architecture, following the idea of goal-conditioned imitation learning, to generate the trajectory learning policy on the skill level offline. Through experimental validations, we corroborate that the proposed learning-based framework can generate high-quality trajectories and find suitable force control policies which adapt to the tasks’ force requirements more efficiently.

Most industrial robots are provided by the manufacturer with a controller that cannot be modified by the user (e.g. a standard PID). This arrangement is commonly referred to as closed control architecture, since it is not possible to program arbitrary control laws. For the implementation of novel algorithms, it is on the contrary necessary to employ an open control architecture, which allows programming any control scheme. For that reason, it is customary to have testbeds that are made up of robot manipulators specially designed for this goal. Another disadvantage of the closed control architecture is that the controller provided by the manufacturer usually does not include a force control term since it allows only to program desired motion trajectories. To overcome these drawbacks without physically modifying the closed control architecture, this contribution presents a novel approach to simultaneously follow position and force trajectories by employing only motion planning, i.e. only by choosing the desired position trajectory. The approach is especially well suited for DC motor actuators with large gear reduction ratios as those of many industrial robots. The convergence of the manipulator position and applied force depends exclusively on the performance of the controller provided by the manufacturer. The approach is tested on a dual-arm cooperative manipulation system made up of two ABB IRB-2400 industrial robots with closed control architecture.

In the context of dense urban environments and climate change, pedestrians’ thermal experience plays an increasingly significant role in people’s health and well-being. In this research, the authors combine the fields of architecture, climate-responsive design, and robotic fabrication with the goal of investigating strategies to improve outdoor thermal comfort for pedestrians in cities with frequent extreme heat events. Based on a case study in the city of Munich, this paper presents findings into the technological approaches and methods for location-specific climate-resilient brick facades using robotic assembly. To achieve this goal, different bricklaying patterns were investigated to create a self-shading effect and thus reduce solar radiation and ultimately achieve an improved thermal condition for pedestrians moving along urban facades at street level. Using computer-aided microclimate simulation, generic self-shading brick pattern designs were tailored to highly location-specific microclimate requirements. Robotic assembly technology was used to produce such tailored, non-standard brickwork facades. The results of this research led to a data-informed design process with a demonstrator object being realized at 1:1 scale with a height of 2 m and a length of 3 m using a collaborative robot on site. Thermal measurements on the built demonstrator provided indications of reduced surface temperatures despite high solar radiation and thus validated the location-specific self-shading effects according to solar radiation simulation.

Complex contact-rich insertion is a ubiquitous robotic manipulation skill and usually involves nonlinear and low-clearance insertion trajectories as well as varying force requirements. A hybrid trajectory and force learning framework can be utilized to generate high-quality trajectories by imitation learning and find suitable force control policies efficiently by reinforcement learning. However, with the mentioned approach, many human demonstrations are necessary to learn several tasks even when those tasks require topologically similar trajectories. Therefore, to reduce human repetitive teaching efforts for new tasks, we present an adaptive imitation framework for robot manipulation. The main contribution of this work is the development of a framework that introduces dynamic movement primitives into a hybrid trajectory and force learning framework to learn a specific class of complex contact-rich insertion tasks based on the trajectory profile of a single task instance belonging to the task class. Through experimental evaluations, we validate that the proposed framework is sample efficient, safer, and generalizes better at learning complex contact-rich insertion tasks on both simulation environments and on real hardware.

The construction sector is investigating wood as a highly sustainable material for fabrication of architectural elements. Several researchers in the field of construction are currently designing novel timber structures as well as novel solutions for fabricating such structures, i.e. robot technologies which allow for automation of a domain dominated by skilled craftsman. In this paper, we present a framework for closing the loop between the design and robotic assembly of timber structures. On one hand, we illustrate an extended automation process that incorporates learning by demonstration to learn and execute a complex assembly of an interlocking wooden joint. On the other hand, we describe a design case study that builds upon the specificity of this process, to achieve new designs of construction elements, which were previously only possible to be assembled by skilled craftsmen. The paper provides an overview of a process with different levels of focus, from the integration of a digital twin to timber joint design and the robotic assembly execution, to the development of a flexible robotic setup and novel assembly procedures for dealing with the complexity of the designed timber joints. We discuss synergistic results on both robotic and construction design innovation, with an outlook on future developments.

The aim of this paper is to present the genetic algorithm used in programming a solar tracker robot. The solar tracker robot can be used to increase the efficiency of solar panels by rotating them toward the Sun. When the efficiency of solar panels is increased, they can reduce the space occupied by them. Efficiency can also be increased by genetic algorithms, where a solar panel can evolve and thus track the Sun more precisely. The genetic algorithm implemented in the solar tracker is the main focus of this paper. The genetic algorithm is presented with a flow chart and formulas. After this, the system is implemented on a solar tracker robot and validated with real experiments. The solar tracker robot algorithm is implemented in the C language on a microcontroller built on an FPGA platform.

Simulation with a reasonable physical model is important to develop control algorithms for robots quickly, accurately, and safely without damaging the associated physical systems in various environments. However, it is difficult to choose the suitable tool for simulating a specific project. To help users in selecting the best tool when simulating a given project, we compare the performance of the four widely used physics engines, namely, ODE, Bullet, Vortex, and MoJoco, for various simple and complex industrial scenarios. We first summarize the technical algorithms implemented in each physics engine. We also designed four simulation scenarios ranging from simple scenarios for which analytic solution exists to complex industrial scenarios to compare the performance of each physics engine. We then present the simulation results in the default settings of all the physics engines, and analyze the behavior and contact force of the simulated objects.

A personal view of some of the more important lessons learned from the module production for the CMS silicon tracker. This work took place from about 2002–2005. The focus is on areas where I had strong personal involvement; therefore, the tasks of hybrid production, hybrid assembly, and the wire bonding of modules and hybrids are emphasized. This article will first give a general description of the silicon tracker project and how the module production was organized. Then, there will be description of several of the key issues or problems during the production and how they were resolved. Some recommendations for future similar large-scale productions will be given.

Peg-in-hole is one of the most frequent mating features between parts, where the large friction resistance and poor contact situations result in the failure of part mating. To address this problem, this paper proposes a screw insertion method in peg-in-hole assembly for axial friction reduction. First, the effect of screw motion on axial friction reduction is analyzed with point contact and face contact. Second, the screw insertion method in clearance-fit peg-in-hole assembly with point contact is discussed; this method works better than the conventional linear insertion method for jamming prevention; the reciprocate screw insertion strategy is also investigated to reduce the influence of axis offset on screw insertion. Third, the screw insertion method in interference-fit peg-in-hole assembly with face contact is discussed. Finally, axial friction reduction is validated with experiments of point contact with screw motion and experiments of screw insertion in clearance-fit and interference-fit peg-in-hole assembly.

Large-aperture space telescope is the key equipment for studying the origin and evolution of the universe and various celestial bodies. Developing large-aperture space telescope is challenging and has become a priority in the field of optical engineering. Due to the mass and volume constraints of launch vehicles, it has to be designed and launched in modules and then assembled on-orbit by space robots. In this paper, we present a conceptual of 10-meter diameter modular space telescope together with robotic assembly strategy. To fulfill on-orbit assembly and maintenance of the space telescope, a novel assembly robot with a ring-shaped mobile base and a redundant stretchable manipulator is proposed. The time-optimal trajectory planning based on genetic algorithm is utilized to achieve efficient assembly. The preliminary ground experiment of assembling submirror modules of space telescope using the KUKA LWR iiwa-7 is carried to verify the feasibility of the proposed method, and the results show that the assembly task of telescope modules can be accomplished successfully with appropriate forces and torques.

A through-glass via (TGV) provides a vertical electrical connection through a glass substrate. TGVs are used in advanced packaging solutions, such as glass interposers and wafer-level packaging of microelectromechanical systems (MEMS). However, TGVs are challenging to realize because via holes in glass typically do not have a sufficiently high-quality sidewall profile for super-conformal electroplating of metal into the via holes. To overcome this problem, we demonstrate here that the via holes can instead be filled by magnetically assembling metal wires into them. This method was used to produce TGVs with a typical resistance of 64 mΩ, which is comparable with other metal TGV types reported in the literature. In contrast to many TGV designs with a hollow center, the proposed TGVs can be more area efficient by allowing solder bump placement directly on top of the TGVs, which was demonstrated here using solder-paste jetting. The magnetic assembly process can be parallelized using an assembly robot, which was found to provide an opportunity for increased wafer-scale assembly speed. The aforementioned qualities of the magnetically assembled TGVs allow the realization of glass interposers and MEMS packages in different thicknesses without the drawbacks associated with the current TGV fabrication methods.

The tight tolerance peg-in-hole process brings great challenges for robotic assembly. Force control-based methods have been proposed to generate complex compliant behavior to deal with the shape and clearance variances. However, existing solutions are based on the assumption that the peg and hole parts are fixed during the assembly process and can absorb the contact force completely. For this purpose, customized fixtures have to be designed and utilized, which greatly affect the system’s deployment cost, time, and flexibility. Considering the fact that in an assembly, the parts are naturally related to each other, this paper studies the irregular-shaped peg-in-hole assembly problem with partial constraints. Firstly, geometric and force model are developed for the natural constraints between the parts; by analyzing the behavior of the partial constraint, a control policy is proposed to compensate the position errors and drive the parts to a stable equilibrium point; For the irregular-shaped parts, a multiple-stage searching method is developed to efficiently search for the real hole location; Finally, a switching force/position hybrid controller is designed to coordinate the alignment, searching and insertion processes. The method is implemented in a real platform. The experiment results verify the effectiveness of the proposed methods.

Assembly tasks executed by a robot have been studied broadly. Robot assembly applications in industry are achievable by a well-structured environment, where the parts to be assembled are located in the working space by fixtures. Recent changes in manufacturing requirements, due to unpredictable demanded products, push the factories to seek new smart solutions that can autonomously recover from failure conditions. In this way, new dual arm robot systems have been studied to design and explore applications based on its dexterity. It promises the possibility to get rid of fixtures in assembly tasks, but using less fixtures increases the uncertainty on the location of the components in the working space. It also increases the possibility of collisions during the assembly sequence. Under these considerations, adding perception such as force/torque sensors have been done to produce useful data to perform control actions. Unfortunately, the interaction forces between mating parts produced non-linear behavior. Consequently, machine learning algorithms have been considered an alternative tool to avoid the non-linearity. In this work we introduce an assembly strategy for an industrial dual arm robot based on the combination of a discrete event controller and Deep Neural Networks (DNN) to solve the peg-in-hole assembly. Our results show the difference between the use of DNN with one and with two force/torque sensors during the assembly task and demonstrate a 30% increase in the assembly success ratio when using a double force/torque sensor.

This paper presents an autonomous robotic assembly system for Soma cube blocks, which, after observing the individual blocks and their assembled shape, quickly plans and executes the assembly motion sequence that picks up each block and incrementally build the target shape. A multi stage planner is used to find the suitable assembly solutions, assembly sequences and grip sequences considering various constraints, and re-grasping is used when the block target pose is not directly realizable or the block pose is ambiguous. The suggested system is implemented for a commercial UR5e robotic arm and a novel two degrees of freedom (DOF) gripper capable of in-hand manipulation, which further speeds up the manipulation speed. It was experimentally validated through a public competitive demonstration, where the suggested system completed all assembly tasks reliably with outstanding performance.

Industrial robot manipulators are playing a significant role in modern manufacturing industries. Though peg-in-hole assembly is a common industrial task that has been extensively researched, safely solving complex, high-precision assembly in an unstructured environment remains an open problem. Reinforcement-learning (RL) methods have proven to be successful in autonomously solving manipulation tasks. However, RL is still not widely adopted in real robotic systems because working with real hardware entails additional challenges, especially when using position-controlled manipulators. The main contribution of this work is a learning-based method to solve peg-in-hole tasks with hole-position uncertainty. We propose the use of an off-policy, model-free reinforcement-learning method, and we bootstraped the training speed by using several transfer-learning techniques (sim2real) and domain randomization. Our proposed learning framework for position-controlled robots was extensively evaluated in contact-rich insertion tasks in a variety of environments.

In manufacturing, traditional task pre-programming methods limit the efficiency of human–robot skill transfer. This paper proposes a novel task-learning strategy, enabling robots to learn skills from human demonstrations flexibly and generalize skills under new task situations. Specifically, we establish a markerless vision capture system to acquire continuous human hand movements and develop a threshold-based heuristic segmentation algorithm to segment the complete movements into different movement primitives (MPs) which encode human hand movements with task-oriented models. For movement primitive learning, we adopt a Gaussian mixture model and Gaussian mixture regression (GMM-GMR) to extract the optimal trajectory encapsulating sufficient human features and utilize dynamical movement primitives (DMPs) to learn for trajectory generalization. In addition, we propose an improved visuo-spatial skill learning (VSL) algorithm to learn goal configurations concerning spatial relationships between task-relevant objects. Only one multioperation demonstration is required for learning, and robots can generalize goal configurations under new task situations following the task execution order from demonstration. A series of peg-in-hole experiments demonstrate that the proposed task-learning strategy can obtain exact pick-and-place points and generate smooth human-like trajectories, verifying the effectiveness of the proposed strategy.

Recently demonstrated robotic assembling technologies for fuel cell stacks used fuel cell components manually pre-arranged in stacks (presenters). Identifying the original orientation of fuel cell components and loading them in presenters for a subsequent automated assembly process is a difficult, repetitive work cycle which if done manually, deceives the advantages offered by either the automated fabrication technologies for fuel cell components or by the robotic assembly processes. We present for the first time a robotic technology which enables the integration of automated fabrication processes for fuel cell components with a robotic assembly process of fuel cell stacks into a fully automated fuel cell manufacturing line. This task uses a Yaskawa Motoman SDA5F dual arm robot with integrated machine vision system. The process is used to identify and grasp randomly placed, slightly asymmetric fuel cell components, to reorient them all in the same position and stack them in presenters in preparation for a subsequent robotic assembly process. The process was demonstrated as part of a larger endeavor of bringing to readiness advanced manufacturing technologies for alternative energy systems, and responds the high priority needs identified by the U.S. Department of Energy for fuel cells manufacturing research and development.

This paper addresses issues related to the design and implementation of robotic assembly tasks. Specifically, we consider automatic assembly systems with real-time visual sensing for the back shells of cellular phones. Typically, industrial assembly tasks are accomplished using either the look-then-move open-loop or the look-and-move closed-loop control approach. For either approach, successful assembly requires that issues concerned with task accuracy must be considered based on camera calibration parameters. For the hand-eye robotic system to operate in real-time, one must adopt an appropriate control structure to maximize task efficiency. Simple and repetitive assembly tasks can be performed quickly through look-then-move open-loop controls. However, relatively slower look-and-move closed-loop control approaches are better suited for complex tasks or those requiring greater flexibility. To accomplish automatic assembly tasks with real-time visual sensing, either eye-to-hand or eye-in-hand vision must be employed. The proposed vision-based control approaches for back shell assembly tasks are likely to have real potential in industrial manufacturing applications.

Learning from demonstration (LfD) has been used to help robots to implement manipulation tasks autonomously, in particular, to learn manipulation behaviors from observing the motion executed by human demonstrators. This paper reviews recent research and development in the field of LfD. The main focus is placed on how to demonstrate the example behaviors to the robot in assembly operations, and how to extract the manipulation features for robot learning and generating imitative behaviors. Diverse metrics are analyzed to evaluate the performance of robot imitation learning. Specifically, the application of LfD in robotic assembly is a focal point in this paper.