Towards single-arm reaching for humanoids in dynamic environments

The problem of humanoid agents or robots reaching to arbitrary targets in environments with static and dynamic obstacles has not yet been investigated in detail. Typical approaches include using randomized motion planning or performing simple planning (i.e., linear interpolation between initial and target positions and orientations of the hand) in operational space, hoping that inter-link and agent-environment collisions do not occur. In this paper, we test the most popular algorithms for motion generation for singlearm reaching in environments with randomly placed obstacles of random sizes. Additionally, an attempt is made to formalize the concept of motor primitives, and a motor primitive implementation is tested in the aforementioned experiments. An analysis of the efficacy of the algorithms for reaching in static environments is conducted, and the extensibility of the algorithms towards reaching in dynamic environments is discussed.