Long-Horizon Manipulation of Unknown Objects via Task and Motion Planning with Estimated Affordances

IEEE International Conference on Robotics and Automation(2022)

引用 30|浏览42
We present a strategy for designing and building very general robot manipulation systems using a general-purpose task-and-motion planner with both engineered and learned modules that estimate properties and affordances of unknown objects. Such systems are closed-loop policies that map from RGB images, depth images, and robot joint encoder measurements to robot joint position commands. We show that this strategy leads to intelligent behaviors even without a priori knowledge regarding the set of objects, their geometries, and their affordances. We show how these modules can be flexibly composed with robot-centric primitives using the PDDLStream task and motion planning framework. Finally, we demonstrate that this strategy can enable a single policy to perform a wide variety of real-world multi-step manipulation tasks, generalizing over a broad class of objects, arrangements, and goals, without prior knowledge of the environment or re-training.
long-horizon manipulation,unknown objects,estimated affordances,general robot manipulation systems,general-purpose task,-motion planner,engineered learned modules,estimate properties,closed-loop policies,RGB images,depth images,robot joint encoder measurements,robot joint position,intelligent behaviors,robot-centric primitives,motion planning framework,single policy,real-world multistep manipulation tasks
AI 理解论文
Chat Paper