Modular robots hold the promise of versatility in that their components can be re-arranged to adapt the robot design to a task at deployment time. Even for the simplest designs, determining the optimal design is exponentially complex due to the number of permutations of ways the modules can be connected. Further, when selecting the design for a given task, there is an additional computational burden in evaluating the capability of each robot, e.g., whether it can reach certain points in the workspace. This work uses deep reinforcement learning to create a search heuristic that allows us to efficiently search the space of modular serial manipulator designs. We show that our algorithm is more computationally efficient in determining robot designs for given tasks in comparison to the current state-of-the-art.

Soft sensors have continued growing interest because they enable both passive conformal contact and provide active contact data from the sensor properties. However, the same properties of conformal contact result in faster deterioration of soft sensors and larger variations in their response characteristics over time and across samples, inhibiting their ability to be long-lasting and replaceable. ReSkin is a tactile soft sensor that leverages machine learning and magnetic sensing to offer a low-cost, diverse and compact solution for long-term use. Magnetic sensing separates the electronic circuitry from the passive interface, making it easier to replace interfaces as they wear out while allowing for a wide variety of form factors. Machine learning allows us to learn sensor response models that are robust to variations across fabrication and time, and our self-supervised learning algorithm enables finer performance enhancement with small, inexpensive data collection procedures. We believe that ReSkin opens the door to more versatile, scalable and inexpensive tactile sensation modules than existing alternatives.

No abstract

Motion planning and control to generate human like walking with bipedal robots has been an active area of research for several decades. Besides walking, humans are capable of performing several other complex maneuvers using their limbs. A handspring is one such maneuver. During this acrobatic move, a person performs a complete revolution of the body by first placing her hands as if in a handstand, then pushing off of the floor with her hands to return to a standing position. Often, acrobats leverage the momentum from one handspring to perform a series of continuous back-to-back handsprings. In this paper, we present handspring trajectory planning for a simple two link robot using numerical optimization. We are able to generate near-periodic perpetual handspring-like maneuvers using two different actuation strategies (i) a constant torque at the hip joint, and (ii) an impulsive actuation to the imminent free leg.

No abstract