Controlling the contact force during interactions is an inherent requirement for locomotion and manipulation tasks. Current reinforcement learning approaches to locomotion and manipulation rely implicitly on forceful interaction to accomplish tasks but do not explicitly regulate it. This paper proposes a reinforcement learning task specification that focuses on matching desired contact force levels. Integrating force control with the coordination of a robot’s body and arm, we present an end-to-end policy for legged manipulator control. Force control enables us to realize compliant gripper and whole-body pulling movements that have not been previously demonstrated using a learned policy. It also facilitates a characterization of the force-tracking performance of learned policies in simulation and the real world, indicating their performance potential for force-critical tasks. Video is available at the project website: https://tif-twirl-13.github.io/learning-compliance.

DribbleBot (Dexterous Ball Manipulation with a Legged Robot) is a legged robotic system that can dribble a soccer ball under the same real-world conditions as humans. We identify key challenges of in-the-wild soccer ball manipulation, including variable ball motion dynamics and perception using body-mounted cameras. To overcome these challenges, we propose a domain and task specification for learning viable soccer dribbling behaviors in simulation that transfer to real fields. Our system provides promising evidence that current legged robots are physically capable and adequately sensorized for varied and dynamic real-world soccer play. Video is available at https://gmargoll.github.io/dribblebot.

We address the problem of enabling quadrupedal robots to perform precise shooting skills in the real world using reinforcement learning. Developing algorithms to enable a legged robot to shoot a soccer ball to a given target is a challenging problem that combines robot motion control and planning into one task. To solve this problem, we need to consider the dynamics limitation and motion stability during the control of a dynamic legged robot. Moreover, we need to consider motion planning to shoot the hard-to-model deformable ball rolling on the ground with uncertain friction to a desired location. In this paper, we propose a hierarchical framework that leverages deep reinforcement learning to train (a) a robust motion control policy that can track arbitrary motions and (b) a planning policy to decide the desired kicking motion to shoot a soccer ball to a target. We deploy the proposed framework on an A1 quadrupedal robot and enable it to accurately shoot the ball to random targets in the real world.

Wearable robots like ankle exoskeletons have demonstrated the capability to enhance human mobility and to reduce biological efforts of human locomotion. The type of assistance provided by ankle exoskeletons could influence the lower leg muscle activities during human walking. This article aimed to systematically evaluate the lower leg muscle activities under different ankle exoskeleton assistance conditions. We measured multiple electromyography-based metrics of five lower leg muscles, while the participants walked with an ankle exoskeleton on a treadmill. Nine assistance conditions, which combined three peak times (46%, 49%, and 52% of stride time) and three peak torque levels (0.3, 0.5, and 0.7 N·m·kg-1), are applied to assist plantarflexion during ankle push-off. Nine healthy subjects participated in the experiments. Of all investigated muscles, the activity level of l.SOL is influenced the most when exoskeleton assistance is applied. The root mean square of l.SOL activity reduces by 33.6 ± 14.0% under one assistance condition compared to walking without the exoskeleton. Our results can be used to guide studies on mechanical and control designs to improve neuromuscular interactions between exoskeletons and wearers.