RATS: A Robotic Arm Training System Designed for Rats

This paper presents a novel robotic system to characterize and retrain reaching in rats. This robot is intended to be a research platform for rehabilitation of forelimb movements in rats. In this paper, we focus on the design of this robotic system. We present the design requirements, mathematical models, and details of the physical device. A parallel mechanism with a special alignment of the component chains is used to accommodate observed reaching motions of a rat's forelimb. Additionally, we demonstrate the use of this robot to record forelimb trajectories. Three healthy rats were used to record repeated reaching motions while the robot applied nearly zero force. We believe that this robotic system can be used in future training studies with rats who have impaired arm motions due to a neurological insult.

Deep learning can accelerate grasp-optimized motion planning

Robots for picking in e-commerce warehouses require rapid computing of efficient and smooth robot arm motions between varying configurations. Recent results integrate grasp analysis with arm motion planning to compute optimal smooth arm motions; however, computation times on the order of tens of seconds dominate motion times. Recent advances in deep learning allow neural networks to quickly compute these motions; however, they lack the precision required to produce kinematically and dynamically feasible motions. While infeasible, the network-computed motions approximate the optimized results. The proposed method warm starts the optimization process by using the approximate motions as a starting point from which the optimizing motion planner refines to an optimized and feasible motion with few iterations. In experiments, the proposed deep learning–based warm-started optimizing motion planner reduces compute and motion time when compared to a sampling-based asymptotically optimal motion planner and an optimizing motion planner. When applied to grasp-optimized motion planning, the results suggest that deep learning can reduce the computation time by two orders of magnitude (300×), from 29 s to 80 ms, making it practical for e-commerce warehouse picking.

RATS: A Robotic Arm Training System Designed for Rats

This paper presents a novel robotic system to characterize and retrain reaching in rats. This robot is intended to be a research platform for rehabilitation of forelimb movements in rats. In this paper, we focus on the design of this robotic system. We present the design requirements, mathematical models, and details of the physical device. A parallel mechanism with a special alignment of the component chains is used to accommodate observed reaching motions of a rat’s forelimb. Additionally, we demonstrate the use of this robot to record forelimb trajectories. Three healthy rats were used to record repeated reaching motions while the robot applied nearly zero force. We believe that this robotic system can be used in future training studies with rats who have impaired arm motions due to a neurological insult.

Analysis of Arm Joint Torques at Ball-Release for Set and Jump Shots in Basketball

A kinetic model of the shooting arms has estimated arm joint torques for one-hand set- and jump-shots in basketball. The dynamic model has three rigid planar links with rotational joints imitating an upper arm, forearm and hand with shoulder, elbow and wrist joints. In general shots controlled by hand, forearm and upper arm motions, there are many torque combinations of shoulder, elbow and wrist joints to produce shooters’ desired ball-release speed, angle and backspin angular velocity. The minimum of the sum of squares of the torque combinations exists at ball-release, when the release angular velocities of the hand and forearm are equal, and the release angular accelerations of the hand and forearm are also equal. Each torque of the shooting arm joints for the set-shot with upward shoulder speed is smaller than that for the jump-shot. Shoulder, elbow and wrist torques increase in proportion to horizontal shot distances. As release backspin angular speed increases, each of the shoulder, elbow and wrist torques also increases. The torques of the shoulder, elbow and hand affect the horizontal shot distance and the ball-release backspin.

A Parallel, 2-DOF Exoskeleton for the Human Shoulder: Device Characterization and Preliminary Results on Healthy Subjects

BackgroundHumans are highly reliant on the efficient function of their upper limb. Neuromuscular disorders that impair the function of the shoulder consequently reduce quality of life. Robotic rehabilitation serves as an attractive treatment choice due to its promising results and its ability to alleviate the demands on therapists and clinicians. Nevertheless, current robotic architectures are not optimized for the human shoulder but are more apt for industrial environments. Pneumatically powered soft robotic actuators present an attractive method to create shoulder exoskeletons due to their compliance and relatively low mass. However, current actuators lack the necessary functions to provide support to the entire shoulder’s range of motion.MethodsA modular, fabric pneumatic actuator was constructed. The actuator design allows it to perform three-dimensional (3-D) bends with minimal resistance. Four actuators were combined to create a soft shoulder exoskeleton. Each actuator drives one direction of motion: elevation and depression, rotation of the plane of elevation. The torque output of the actuator was measured using a customized two-axis torque measurement system. Exoskeleton functionality was tested through surface electromyography of relevant shoulder muscles. 10 healthy subjects were recruited and performed arm motions under the assistance of the exoskeleton.ResultsThe actuator can reach full bending (>360°) with low pressures (~10kPA). Its torque output is highly dependent on its geometry. Moreover, torque output is reduced as the bending angles increase. The actuators installed on the exoskeleton output 11.15N-m of torque at the neutral position, and 4.44 N-m at 90° shoulder elevation. The test on healthy subjects showed that use of the exoskeleton reduces muscle activation by up to 65% when performing shoulder elevation, and up to 34% when rotating the plane of elevation. Use of the exoskeleton also resulted in a change in arm trajectory when performing elevation and depression movements.ConclusionsThe reduction in muscle activation highlights the ability of a soft-robotic exoskeleton in supporting arm movements. Moreover, the presented exoskeleton design successfully demonstrated its ability to generate two degree-of-freedom support for the human shoulder.

Estimating Exerted Hand Force via Force Myography to Interact with a Biaxial Stage in Real-Time by Learning Human Intentions: A Preliminary Investigation

Force myography (FMG) signals can read volumetric changes of muscle movements, while a human participant interacts with the environment. For collaborative activities, FMG signals could potentially provide a viable solution to controlling manipulators. In this paper, a novel method to interact with a two-degree-of-freedom (DoF) system consisting of two perpendicular linear stages using FMG is investigated. The method consists in estimating exerted hand forces in dynamic arm motions of a participant using FMG signals to provide velocity commands to the biaxial stage during interactions. Five different arm motion patterns with increasing complexities, i.e., “x-direction”, “y-direction”, “diagonal”, “square”, and “diamond”, were considered as human intentions to manipulate the stage within its planar workspace. FMG-based force estimation was implemented and evaluated with a support vector regressor (SVR) and a kernel ridge regressor (KRR). Real-time assessments, where 10 healthy participants were asked to interact with the biaxial stage by exerted hand forces in the five intended arm motions mentioned above, were conducted. Both the SVR and the KRR obtained higher estimation accuracies of 90–94% during interactions with simple arm motions (x-direction and y-direction), while for complex arm motions (diagonal, square, and diamond) the notable accuracies of 82–89% supported the viability of the FMG-based interactive control.