ProDapt: Proprioceptive Adaptation using Long-term Memory Diffusion
This work addresses the need for robust robot operation in sensor-limited environments like military or space applications, representing an incremental improvement by adding long-term memory to existing diffusion-based imitation learning methods.
The paper tackles the problem of robots relying on exteroceptive sensors like cameras, which can fail in challenging environments such as space or underwater, by proposing ProDapt, a method that incorporates long-term memory of past contacts into diffusion models to enable task completion using only proprioceptive data, demonstrated with a UR10e robotic arm in simulations and real experiments.
Diffusion models have revolutionized imitation learning, allowing robots to replicate complex behaviours. However, diffusion often relies on cameras and other exteroceptive sensors to observe the environment and lacks long-term memory. In space, military, and underwater applications, robots must be highly robust to failures in exteroceptive sensors, operating using only proprioceptive information. In this paper, we propose ProDapt, a method of incorporating long-term memory of previous contacts between the robot and the environment in the diffusion process, allowing it to complete tasks using only proprioceptive data. This is achieved by identifying "keypoints", essential past observations maintained as inputs to the policy. We test our approach using a UR10e robotic arm in both simulation and real experiments and demonstrate the necessity of this long-term memory for task completion.