D-LGP: Dynamic Logic-Geometric Program for Reactive Task and Motion Planning
ICRA 2024
ICRA 2024
Many real-world sequential manipulation tasks involve a combination of discrete symbolic search and continuous motion planning, collectively known as combined task and motion planning (TAMP). However, prevailing methods often struggle with the computational burden and intricate combinatorial challenges, limiting their applications for online replanning in the real world. To address this, we propose Dynamic Logic-Geometric Program (D-LGP), a novel approach integrating Dynamic Tree Search and global optimization for efficient hybrid planning. Through empirical evaluation on three benchmarks, we demonstrate the efficacy of our approach, showcasing superior performance in comparison to state-of-the-art techniques. We validate our approach through simulation and demonstrate its reactive capability to cope with online uncertainty and external disturbances in the real world.
OP-X: Grasp the red block surrounded by X-1 taller blocks.
Grasp-6
Grasp-9
Grasp-12
Grasp-15
Tower-X: Construct a tower in a specific order with X blocks.
Tower-4
Tower-5
Tower-6
Tower-8
Tower-tool-4: Tower construction task with tool usage if blocks are beyond robot's reach.
Real robot experiments
The submitted video