A robotic task planning method
By generating causal dependency subtask graphs through a causal task knowledge base and reverse derivation, the problem of insufficient causal constraints in existing robot task planning is solved. This improves the robot's task understanding and collaborative execution capabilities in complex environments, and enhances the stability of task plans and the efficiency of anomaly repair.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- GUOHAO CARBON (BEIJING) ENERGY TECH RES INST CO LTD
- Filing Date
- 2026-06-11
- Publication Date
- 2026-07-17
AI Technical Summary
Existing robot task planning methods rely on preset action sequences or directly generate action flows, making it difficult to identify the real preconditions and execution consequences between subtasks based on the target state. This results in a lack of causal constraints in task decomposition and difficulty in locating and repairing affected task paths when execution is abnormal, making it difficult to adapt to the frequent model changes, dynamic scheduling, and autonomous decision-making requirements in flexible manufacturing.
A robot task planning method based on causal task knowledge base and causal back-reasoning is adopted to generate a causal dependency subtask graph, and local replanning is performed through execution feedback to improve the causal consistency of task planning and the efficiency of anomaly repair.
It enhances task understanding, generalization adaptation, and collaborative execution capabilities in complex manufacturing environments, improves the stability of task planning and replanning efficiency, and reduces the disturbance to recovery when execution anomalies occur.
Smart Images
Figure CN122401436A_ABST