MIT researchers develop algorithm to optimize sensor placement in soft robots

MIT CSAIL researchers developed a deep-learning algorithm that optimizes sensor placement in soft robots, as reported by MIT News on March 22, 2021. The algorithm outperformed human intuition in simulations, and the research was funded by the National Science Foundation and the Fannie and John Hertz Foundation.

For researchers designing soft robots, automating the process of deciding where to place a limited number of sensors on a robot’s body is important. The MIT researchers say their method selected sensor placements better than human judgment in simulations of three placement configurations for tasks such as grasping.