Randomized Physics-Based Motion Planning for Grasping in Cluttered and Uncertain Environments

Abstract

Abstract-Planning motions to grasp an object in cluttered and uncertain environments is a challenging task, particularly when a collision-free trajectory does not exist and objects obstructing the way are required to be carefully grasped and moved out. This paper takes a different approach and proposes to address this problem by using a randomized physics-based motion planner that permits robot-object and object-object interactions. The main idea is to avoid an explicit high-level reasoning of the task by providing the motion planner with a physics engine to evaluate possible complex multi-body dynamical interactions. The approach is able to solve the problem in complex scenarios, also considering uncertainty in the objects' pose and in the contact dynamics. The work enhances the state validity checker, the control sampler and the tree exploration strategy of a kinodynamic motion planner called KPIECE. The enhanced algorithm, called p-KPIECE, has been validated in simulation and with real experiments. The results have been compared with an ontological physics-based motion planner and with task and motion planning approaches, resulting in a significant improvement in terms of planning time, success rate and quality of the solution path.

PMK—A Knowledge Processing Framework for Autonomous Robotics Perception and Manipulation

“…At the reasoning process level, they do not fully cover the area of TAMP to facilitate the planning process. For example, in motion planning, to deal with rigid bodies in cluttered environments, there is the need to define the way to apply actions such as push/pull, requiring a rich semantic description to be fed to the planner, like the physics-based motion planner in [10]. In task planning, the robot needs to reason on: (a) the feasibility of an action at some instant of the manipulation planning process (according to object features, the the state of the object could change and be out of the robot capabilities, e.g., if the cup is empty the state is graspable and if full the state is pushable), (b) the selection of the placement where the robot must place the object, and (c) the current constraints.…”

“…Sometimes in cluttered environments no collision-free motions exist to move the robot arm to a grasping pose, although a path toward the goal can be found if interactions with movable obstacles are allowed (i.e., the robot clears the path by pushing the obstacles away). This is done using physics-based motion planning strategies such as [10]. Briefly, physics-based motion planning is the evolved form of kinodynamic motion planning, that while planning do not preclude collisions with some obstacles and considers both kinodynamic constraints (such as joint limits and bound over velocities and forces) and physics-based constraints (such as friction and gravity).…”

“…These components play a significant role for bi-manual robot tasks or for multi-robot cooperation. For instance, it could be useful in planners such as [9] that require some geometric reasoning, or in physics-based planners like [10] that require to query about how interaction with the objects is to be done. Also, a knowledge-based sensing module could be integrated with TAMP contingency planners coping with uncertain scenarios, that require reasoning about the sensing system, the features of the environment entities, and sensor limitations.…”

Real-Time Online Re-Planning for Grasping Under Clutter and Uncertainty

“…Moreover, since these are optimization-based methods, even when they are used with a small time limit, they can still output an improved lower-cost trajectory, even if the trajectory is not necessarily reaching a goal state. In contrast, sampling-based planners such as RRTs and PRMs [3], [5] typically do not return a useful solution unless they are run until a path to the goal is found, which can take minutes. Table d i…”

“…Existing work addresses this problem using motion planning followed by open-loop execution [2]- [5] i.e. the robot executes a sequence of actions one after the other without getting any feedback from the environment.…”