Multi-directional search with goal switching for robot path planning

“…The authors suggest a machine-learning based method to estimate the heuristic online. The approach in [12] uses the A* algorithm with an adapted heuristic to approach the physical goal configuration that is closest to the current configuration. Similarly, adaptive A* is used in [13] to identify the best path to traverse multiple goal locations.…”

Section: Introductionmentioning

confidence: 99%

Monte-Carlo Tree Search with Prioritized Node Expansion for Multi-Goal Task Planning

Pfeiffer¹,

Edgar²,

Pham³

2023

Preprint

View full text Add to dashboard Cite

Task planning for robots is computationally challenging due to the combinatorial complexity of the possible action space. This fact is amplified if there are several subgoals to be achieved due to the increased length of the action sequences. In this work, we propose a multi-goal task planning algorithm for deterministic decision processes based on Monte Carlo Tree Search. We augment the algorithm by prioritized node expansion which prioritizes nodes that already have fulfilled some sub-goals. Due to its linear complexity in the number of sub-goals our algorithm is able to identify action sequences of 145 elements to reach the desired goal state with up to 48 sub-goals while the search tree is limited to under 6500 nodes. We use action reduction based on a kinematic reachability criterion to further ease computational complexity. We combine our algorithm with object localization and motion planning and apply it to a real-robot demonstration with two manipulators in an industrial bearing inspection setting.

show abstract

“…By using an optimal discretization method based on a predefined Cartesian value, smaller search spaces with an unchanged solution accuracy can be received [Henrich98e]. By multi-directional search with goal switching, the planning system is enabled to select one of the preferable goal configuration by itself [Henrich98c]. By running a parallel version of the path planner on a workstation cluster with 9 PCs, the planning times can be reduced to only a few seconds [Henrich98a].…”

Section: Basic Approachmentioning

confidence: 99%

On-line path planning by heuristic hierarchical search

Henrich¹,

Wurll²,

Wörn³

IECON '98. Proceedings of the 24th Annual Conference of the IEEE Industrial Electronics Society (Cat. No.98CH36200)

View full text Add to dashboard Cite

In this paper, the problem of path planning for robot manipulators with six degrees of freedom in an on-line provided threedimensional environment is investigated. As a basic approach, the best-first algorithm is used to search in the implicit descrete configuration space. Collisions are detected in the Cartesian workspace by hierarchical distance computation based on the given CAD model. The basic approach is extended by three simple mechanisms and results in a heuristic hierarchical search. This is done by adjusting the stepsize of the search to the distance between the robot and the obstacles. As a first step, we show encouraging experimental results with two degrees of freedom for five typical benchmark problems.

show abstract

Multi-directional search with goal switching for robot path planning

Cited by 11 publications

References 8 publications

Point‐to‐point and multi‐goal path planning for industrial robots

Point‐to‐point and multi‐goal path planning for industrial robots

Monte-Carlo Tree Search with Prioritized Node Expansion for Multi-Goal Task Planning

On-line path planning by heuristic hierarchical search

Contact Info

Product

Resources

About