Mobile Manipulation in Unstructured Environments

Unstructured human environments present a substantial challenge to effective robotic operation. Mobile manipulation in typical human environments requires dealing with novel unknown objects, cluttered workspaces, and noisy sensor data. We present an approach to mobile manipulation in such environments using a combination of 2D and 3D visual processing, tactile and proprioceptive sensor data, fast motion planning, reactive control and monitoring, and reactive grasping. Our approach allows a robot to perform the full range of operations required for mobile manipulation. We demonstrate our approach by using a two-arm mobile manipulation system to pick and place objects. Our approach attempts to maximize safety and robustness by continuously monitoring the task using visual sensors and replanning or aborting the task if necessary. Furthermore, reactive components attempt to correct the desired plan when presented with unexpected information from the changing environment.