Task and Motion Planning for Humanoid Loco-manipulation

-- This work presents an optimization-based task and motion planning (T AMP) framework that unifies planning for locomotion and manipulation through a shared representation of contact modes. We define symbolic actions as contact mode changes, grounding high-level planning in low-level motion. This enables a unified search that spans task, contact, and motion planning while incorporating whole-body dynamics, as well as all constraints between the robot, the manipulated object, and the environment. Results on a humanoid platform show that our method can generate a broad range of physically consistent loco-manipulation behaviors over long action sequences requiring complex reasoning. T o the best of our knowledge, this is the first work that enables the resolution of an integrated T AMP formulation with fully acyclic planning and whole body dynamics with actuation constraints for the humanoid loco-manipulation problem. To perform any reasonable real-world task, a humanoid robot needs to plan whole-body loco-manipulation movements. Solving a holistic optimal control problem with contact constraints has been shown to successfully generate loco-manipulation behaviors in simulation [1], [2]. However, to achieve this these approaches relax complementarity constraints imposed by contact which introduces several artefacts and non-physical behaviors (e.g., force at a distance) in the generated trajectories, all the while being prone to poor local minima due to the non-convexity of the relaxed contact constraints. Furthermore, when multiple objects with sparse contact points (grabbing a handle to move an object or opening a door) are included, they struggle to find a solution.