We present an online learning framework that enables a bimanual robot to acquire diverse juggling patterns directly on physical hardware within minutes, even with a significant sim2real gap. One of the most important lessons from this work is that a model, even when far from reality, can be extremely useful for learning. This motivates a central philosophy of our approach: learning should build upon the robot's current knowledge rather than replace it. Our regularized memory-based learning puts this principle into practice by learning a local model from accumulated experience while retaining the global prior model to extrapolate where experience is sparse. This enables efficient and stable online learning from each new experience without resorting to uninformed exploration over a vast space of possible behaviors. Equally important to continual on-robot learning is safety, allowing the robot to repeatedly practice and improve in the real world. We construct a mutually reachable set that allows safe transitions between successive throws and catches, without driving either arm into a state from which its next action would require violating the robot's joint or actuator limits. Together, these ideas enable a bimanual robot with multi-fingered hands and onboard vision to safely learn and compose five canonical three-ball juggling patterns, including cascade, tennis, half-shower, shower, and box, within less than 5 minutes of real-world interaction. More broadly, this work points toward robots that build upon imperfect prior knowledge and continually refine their behavior through their own real-world experience.

This chapter is about the fundamentals of fabrication, control, and human-robot interaction of a new type of collaborative robotic manipulators, called malleable robots, which are based on adjustable architectures of varying stiffness for achieving high dexterity with lower mobility arms. Collaborative robots, or cobots, commonly integrate six or more degrees of freedom (DOF) in a serial arm in order to allow positioning in constrained spaces and adaptability across tasks. Increasing the dexterity of robotic arms has been indeed traditionally accomplished by increasing the number of degrees of freedom of the system; however, once a robotic task has been established (e.g., a pick-and-place operation), the motion of the end-effector can be normally achieved using less than 6-DOF (i.e., lower mobility). The aim of malleable robots is to close the technological gap that separates current cobots from achieving flexible, accessible manufacturing automation with a reduced number of actuators.

Myoelectric prosthetic hands are typically controlled to move between discrete positions and do not provide sensory feedback to the user. In this work, we present and evaluate a closed-loop, continuous myoelectric prosthetic hand controller, that can continuously control the position of multiple degrees of freedom of a prosthesis while rendering proprioceptive feedback to the user via a haptic feedback armband. Twenty-eight participants without and ten participants with upper limb difference (ULD) were recruited to holistically evaluate the physical and psychological effects of the controller via isolated control and sensory tasks, dexterity assessments, embodiment and task load questionnaires, and post-study interviews. The combination of proprioceptive feedback and continuous control enabled more accurate position and force modulation than without proprioceptive feedback, and restored blindfolded object identification ability to open-loop discrete controller levels. Dexterity assessment and embodiment questionnaire results revealed no significant physical performance or psychological embodiment differences between control types, with the exception of perceived sensation questions, which were significantly higher (p < 0.001) for closed-loop controllers. Key differences between participants with and without ULD were identified, including increasingly lower perceived body completeness and heterogeneity in frustration in participants with ULD, which can inform future development and rehabilitation.

Despite decades of research and development, myoelectric prosthetic hands lack functionality and are often rejected by users. This lack in functionality can be partially attributed to the widely accepted anthropomorphic design ideology in the field; attempting to replicate human hand form and function despite severe limitations in control and sensing technology. Instead, prosthetic hands can be tailored to perform specific tasks without increasing complexity by shedding the constraints of anthropomorphism. In this paper, we develop and evaluate four open-source modular non-humanoid devices to perform the motion required to replicate human flicking motion and to twist a screwdriver, and the functionality required to pick and place flat objects and to cut paper. Experimental results from these devices demonstrate that, versus a humanoid prosthesis, non-humanoid prosthesis design dramatically improves task performance, reduces user compensatory movement, and reduces task load. Case studies with two end users demonstrate the translational benefits of this research. We found that special attention should be paid to monitoring end-user task load to ensure positive rehabilitation outcomes.

End-to-end robot learning, particularly for long-horizon tasks, often results in unpredictable outcomes and poor generalization. To address these challenges, we propose a novel Therblig-Based Backbone Framework (TBBF) as a fundamental structure to enhance interpretability, data efficiency, and generalization in robotic systems. TBBF utilizes expert demonstrations to enable therblig-level task decomposition, facilitate efficient action-object mapping, and generate adaptive trajectories for new scenarios. The approach consists of two stages: offline training and online testing. During the offline training stage, we developed the Meta-RGate SynerFusion (MGSF) network for accurate therblig segmentation across various tasks. In the online testing stage, after a one-shot demonstration of a new task is collected, our MGSF network extracts high-level knowledge, which is then encoded into the image using Action Registration (ActionREG) . Additionally, Large Language Model (LLM)-Alignment Policy for Visual Correction (LAP-VC) is employed to ensure precise action registration, facilitating trajectory transfer in novel robot scenarios. Experimental results validate these methods, achieving 94.37% recall in therblig segmentation and success rates of 94.4% and 80% in real-world online robot testing for simple and complex scenarios, respectively. Supplementary material is available at: https://sites.google.com/view/therbligsbasedbackbone/home

This paper introduces a Cosserat rod based mathematical model for modeling a self-controllable variable curvature soft continuum robot. This soft continuum robot has a hollow inner channel and was developed with the ability to perform variable curvature utilizing a growing spine. The growing spine is able to grow and retract while modifies its stiffness through milli-size particle (glass bubble) granular jamming. This soft continuum robot can then perform continuous curvature variation, unlike previous approaches whose curvature variation is discrete and depends on the number of locking mechanisms or manual configurations. The robot poses an emergent modeling problem due to the variable stiffness growing spine which is addressed in this paper. We investigate the property of growing spine stiffness and incorporate it into the Cosserat rod model by implementing a combined stiffness approach. We conduct experiments with the soft continuum robot in various configurations and compared the results with our developed mathematical model. The results show that the mathematical model based on the adapted Cosserat rod matches the experimental results with only a 3.3% error with respect to the length of the soft continuum robot.

Approaching robotic cloth manipulation using reinforcement learning based on visual feedback is appealing as robot perception and control can be learned simultaneously. However, major challenges result due to the intricate dynamics of cloth and the high dimensionality of the corresponding states, what shadows the practicality of the idea. To tackle these issues, we propose TraKDis , a novel Tra nsformer-based K nowledge Dis tillation approach that decomposes the visual reinforcement learning problem into two distinct stages. In the first stage, a privileged agent is trained, which possesses complete knowledge of the cloth state information. This privileged agent acts as a teacher, providing valuable guidance and training signals for subsequent stages. The second stage involves a knowledge distillation procedure, where the knowledge acquired by the privileged agent is transferred to a vision-based agent by leveraging pre-trained state estimation and weight initialization. TraKDis demonstrates better performance when compared to state-of-the-art RL techniques, showing a higher performance of 21.9%, 13.8%, and 8.3% in cloth folding tasks in simulation. Furthermore, to validate robustness, we evaluate the agent in a noisy environment; the results indicate its ability to handle and adapt to environmental uncertainties effectively. Real robot experiments are also conducted to showcase the efficiency of our method in real-world scenarios.

Human hands are able to grasp a wide range of object sizes, shapes, and weights, achieved via reshaping and altering their apparent grasping stiffness between compliant power and rigid precision. Achieving similar versatility in robotic hands remains a challenge, which has often been addressed by adding extra controllable degrees of freedom, tactile sensors, or specialised extra grasping hardware, at the cost of control complexity and robustness. We introduce a novel reconfigurable four-fingered two-actuator underactuated gripper—the Hydra Hand—that switches between compliant power and rigid precision grasps using a single motor, while generating grasps via a single hydraulic actuator—exhibiting adaptive grasping between finger pairs, enabling the power grasping of two objects simultaneously. The mode switching mechanism and the hand's kinematics are presented and analysed, and performance is tested on two grasping benchmarks: one focused on rigid objects, and the other on items of clothing. The Hydra Hand is shown to excel at grasping large and irregular objects, and small objects with its respective compliant power and rigid precision configurations. The hand's versatility is then showcased by executing the challenging manipulation task of safely grasping and placing a bunch of grapes, and then plucking a single grape from the bunch.

Object recognition and object pose estimation in robotic grasping continue to be significant challenges, since building a labelled dataset can be time consuming and financially costly in terms of data collection and annotation. In this work, we propose a synthetic data generation method that minimizes human intervention and makes downstream image segmentation algorithms more robust by combining a generated synthetic dataset with a smaller real-world dataset (hybrid dataset). Annotation experiments show that the proposed synthetic scene generation can diminish labelling time dramatically. RGB image segmentation is trained with hybrid dataset and combined with depth information to produce pixel-to-point correspondence of individual segmented objects. The object to grasp is then determined by the confidence score of the segmentation algorithm. Pick-and-place experiments demonstrate that segmentation trained on our hybrid dataset (98.9%, 70%) outperforms the real dataset and a publicly available dataset by (6.7%, 18.8%) and (2.8%, 10%) in terms of labelling and grasping success rate, respectively. Supplementary material is available at https://sites.google.com/view/synthetic-dataset-generation.

The manipulation of garments poses research challenges due to their deformable nature and the extensive variability in shapes and sizes. Despite numerous attempts by researchers to address these via approaches involving robot perception and control, there has been a relatively limited interest in resolving it through the co-development of robot hardware. Consequently, the majority of studies employ off-the-shelf grippers in conjunction with dual robot arms to enable bimanual manipulation and high dexterity. However, this dual-arm system increases the overall cost of the robotic system as well as its control complexity in order to tackle robot collisions and other robot coordination issues. As an alternative approach, we propose to enable bimanual cloth manipulation using a single robot arm via novel end effector design—sharing dexterity skills between manipulator and gripper rather than relying entirely on robot arm coordination. To this end, we introduce a new gripper, called G.O.G. , based on a gripper-on-gripper structure where the first gripper independently regulates the span, up to 500mm, between its fingers which are in turn also grippers. These finger grippers consist of a variable friction module that enables two grasping modes: firm and sliding grasps. Household item and cloth object benchmarks are employed to evaluate the performance of the proposed design, encompassing both experiments on the gripper design itself and on cloth manipulation. Experimental results demonstrate the potential of the introduced ideas to undertake a range of bimanual cloth manipulation tasks with a single robot arm. Supplementary material is available at https://sites.google.com/view/gripperongripper .

The application of mechanical and other physical properties to the development of robotic systems that can easily adapt to changing external situations is known as mechanical intelligence. Following this concept, many robot hand designs can produce self-adaptive and versatile grasps with simple underactuated fingers and open-loop control, while mechanical-intelligent strategies for dexterous manipulation are still limited. This paper proposes a mechanical-intelligent technique to facilitate dexterous manipulation, in particular prehensile in-hand manipulation. The proposed strategy is based on the generation of complex spatial trajectories of the hand-object system, controlled in open loop with the minimum number of actuators and using simple low-level non-position modes. This approach is exemplified by the rigorous analysis and testing of a three-fingered two-actuator underactuated robot hand, called the helical hand, which is capable of generating helical prehensile in-hand manipulation of diversiform objects under error tolerance controlled by constant speed algorithm.

When performing cloth-related tasks, such as garment hanging, it is often important to identify and grasp certain structural regions—a shirt's collar as opposed to its sleeve, for instance. However, due to cloth deformability, these manipulation activities, which are essential in domestic, health care, and industrial contexts, remain challenging for robots. In this paper, we focus on how to segment and grasp structural regions of clothes to enable manipulation tasks, using hanging tasks as case study. To this end, a neural network-based perception system is proposed to segment a shirt's collar from areas that represent the rest of the scene in a depth image. With a 10-minute video of a human manipulating shirts to train it, our perception system is capable of generalizing to other shirts regardless of texture as well as to other types of collared garments. A novel grasping strategy is then proposed based on the segmentation to determine grasping pose. Experiments demonstrate that our proposed grasping strategy achieves 92%, 80%, and 50% grasping success rates with one folded garment, one crumpled garment and three crumpled garments, respectively. Our grasping strategy performs considerably better than tested baselines that do not take into account the structural nature of the garments. With the proposed region segmentation and grasping strategy, challenging garment hanging tasks are successfully implemented using an open-loop control policy. Supplementary material is available at https://sites.google.com/view/garment-hanging

Achieving successful robotic manipulation is an essential step towards robots being widely used in industry and home settings. Recently, many learning-based methods have been proposed to tackle this challenge, with imitation learning showing great promise. However, imperfect demonstrations and a lack of feedback from teleoperation systems may lead to poor or even unsafe results. In this work we explore the effect of demonstrator force feedback on imitation learning, using a feedback glove and a robot arm to render fingertip-level and palm-level forces, respectively. 10 participants recorded 5 demonstrations of a pick-and-place task with 3 grippers, under conditions with no force feedback, fingertip force feedback, and fingertip and palm force feedback. Results show that force feedback significantly reduces demonstrator fingertip and palm forces, leads to a lower variation in demonstrator forces, and recorded trajectories that are quicker to execute. Using behavioral cloning, we find that agents trained to imitate these trajectories mirror these benefits, even though agents have no force data shown to them during training. We conclude that immersive demonstrations, achieved with force feedback, may be the key to unlocking safer, quicker-to-execute dexterous manipulation policies.

Through the implementation of reconfigurability to achieve flexibility and adaptation to tasks by morphology changes rather than by increasing the number of joints, malleable robots present advantages over traditional serial robot arms in regards to reduced weight, size, and cost. While limited in degrees of freedom (DOF), malleable robots still provide versatility across operations typically served by systems using higher DOF than required by the tasks. In this article, we present the creation of a 2-DOF malleable robot, detailing the design of joints and malleable link, along with its modeling through forward and inverse kinematics, and a reconfiguration methodology that informs morphology changes based on end effector location-determining how the user should reshape the robot to enable a task previously unattainable. The recalibration and motion planning for making robot motion possible after reconfiguration are also discussed, and thorough experiments with the prototype to evaluate accuracy and reliability of the system are presented. Results validate the approach and pave the way for further research in the area.

Traditional dynamic models of continuum robots are in general computationally expensive and not suitable for real-time control. Recent approaches using learning-based methods to approximate the dynamic model of continuum robots for control have been promising, although real data hungry -- which may cause potential damage to robots and be time consuming -- and getting poorer performance when trained with simulation data only. This paper presents a model-based learning framework for continuum robot closed-loop control that, by combining simulation and real data, shows to require only 100 real data to outperform a real-data-only controller trained using up to 10000 points. The introduced data-efficient framework with three control policies has utilized a Gaussian process regression (GPR) and a recurrent neural network (RNN). Control policy A uses a GPR model and a RNN trained in simulation to optimize control outputs for simulated targets; control policy B retrains the RNN in policy A with data generated from the GPR model to adapt to real robot physics; control policy C utilizes policy A and B to form a hybrid policy. Using a continuum robot with soft spines, we show that our approach provides an efficient framework to bridge the sim-to-real gap in model-based learning for continuum robots.

Using a delta-manipulator for stabilisation of an end-effector to perform precise spatial positioning is a current area of interest in aerial manipulation. High speed precision movements of a manipulator can cause disturbances to the aerial platform, which hinders trajectory tracking and in some cases could be sufficient to cause a loss of control of the vehicle. In this paper, a statically balanced delta aerial manipulator is developed and evaluated. The system is balanced using three counter-masses to reduce the force imparted onto the base and thus reduce perturbations to the movement of the drone. The system is thoroughly tested following trajectories while mounted to a force sensor and while on-board an aerial vehicle. Results show that the forces transmitted to the base in all axes are reduced considerably, however improvements in overall flight accuracy are not observed in aerial settings. Design lessons to make a balanced delta-manipulator viable for practical implementation on an aerial vehicle are discussed in depth. A video summarising the flight testing results is available at https://youtu.be/fXKnosnVKCk.

This paper introduces an approach for solving the in-hand manipulation problem, that is, the change of a grasped object pose from an initial configuration to a final one without breaking contact, with robot fingers having out-of-plane motion, redundancy, and equipped with soft fingertips. The proposed technique is based on keeping the initial grasp equilibrium condition as a constraint to resolve finger redundancy. Two important aspects of in-hand manipulation with soft fingertips are then studied; namely: i) the modeling of soft fingertip contacts between fingers and 3D objects, and ii) an efficient method for computing joint angles to move a grasped object between different poses without losing grasp equilibrium. Numerical and empirical experiments using soft fingertips with different shore hardnesses with a two-fingered robot hand, composed of four-degree-of-freedom fingers with out-of-plane motion, are conducted. Results successfully validate the introduced strategy and its components.

Malleable robots are a type of reconfigurable serial robot capable of adapting their topology, through the use of variable stiffness malleable links, to desired tasks and workspaces by varying the relative positioning between their revolute joints. However, their reconfiguration is nontrivial, lacking intuitive communication between the human and the robot, and a method of efficiently aligning the end effector to a desired position. In this article, we present the design of an interactive augmented reality (AR) alignment interface, which helps a malleable robot understand the user’s task requirements, visualizes to the user the requested robot’s configuration and its workspace, and guides the user in reconfiguring the robot to achieve that configuration. Through motion tracking of a physical two degree-of-freedom (2 DoF) malleable robot, which can achieve an infinite number of workspaces, we compute the accuracy of the system in terms of initial calibration and overall accuracy, and demonstrate its viability. The results demonstrated a good performance, with an average repositioning accuracy of 9.64 ± 2.06 mm and an average base alignment accuracy of 10.54 ± 4.32 mm in an environment the size of 2,000 mm × 2,000 mm × 1,200 mm.

Autonomous grasping of novel objects that are previously unseen to a robot is an ongoing challenge in robotic manipulation. In the last decades, many approaches have been presented to address this problem for specific robot hands. The UniGrasp framework, introduced recently, has the ability to generalize to different types of robotic grippers; however, this method does not work on grippers with closed-loop constraints and is data-inefficient when applied to robot hands with multi-grasp configurations. In this letter, we present EfficientGrasp, a generalized grasp synthesis and gripper control method that is independent of gripper model specifications. EfficientGrasp utilizes a gripper workspace feature rather than UniGrasp’s gripper attribute inputs. This reduces memory use by 81.7% during training and makes it possible to generalize to more types of grippers, such as grippers with closed-loop constraints. The effectiveness of EfficientGrasp is evaluated by conducting object grasping experiments both in simulation and real-world; results show that the proposed method also outperforms UniGrasp when considering only grippers without closed-loop constraints. In these cases, EfficientGrasp shows 9.85% higher accuracy in generating contact points and 3.10% higher grasping success rate in simulation. The real-world experiments are conducted with a gripper with closed-loop constraints, which UniGrasp fails to handle while EfficientGrasp achieves a success rate of 83.3%. The main causes of grasping failures of the proposed method are analyzed, highlighting ways of enhancing grasp performance.

Prosthetic hand control research typically focuses on developing increasingly complex controllers to achieve diminishing returns in pattern recognition of muscle activity signals, making models less suitable for user calibration. Some works have investigated transfer learning to alleviate this, but such approaches increase model size dramatically-thus reducing their suitability for implementation on real prostheses. In this work, we propose a novel, non-parametric controller that uses the Wasserstein distance to compare the distribution of incoming signals to those of a set of reference distributions, with the intended action classified as the closest distribution. This controller requires only a single capture of data per reference distribution, making calibration almost instantaneous. Preliminary experiments building a reference library show that, in theory, users are able to produce up to 9 distinguishable muscle activity patterns. However, in practice, variation when repeating actions reduces this. Controller accuracy results show that 10 non-disabled and 1 disabled participant were able to use the controller with a maximum of two recalibrations to perform 6 actions at an average accuracy of 89.9% and 86.7% respectively. Practical experiments show that the controller allows users to complete all tasks of the Jebsen-Taylor Hand Function Test, although the task of picking and placing small common objects required on average more time than the protocol’s maximum time.