Body-aware multitask planning method and device, electronic equipment and storage medium
By generating parallel-executable task combinations through instance database parsing and resource status detection, and combining real-time sensing data for action instruction planning, the problem of resource contention and spatial conflict in multi-task execution of embodied sensing devices is solved, achieving efficient and stable multi-task execution.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- 深圳若愚科技有限公司
- Filing Date
- 2026-05-08
- Publication Date
- 2026-07-24
AI Technical Summary
Existing multi-task execution planning methods for embodied sensing devices lack systematic joint judgment, which may lead to resource contention, state mutual exclusion, and execution space conflicts between tasks, affecting execution efficiency and stability.
The task instructions are instantiated, parsed, and prioritized using an instance database. Conflict detection is performed in conjunction with hardware and environmental resource status to generate a combination of tasks that can be executed in parallel. Real-time perception data and execution feedback are introduced to plan action instructions and form a multi-threaded action sequence.
It improves the reliability and overall execution efficiency of multi-task planning, enhances the adaptability to complex scene changes, avoids task conflicts and resource interference, and improves the coordinated utilization of equipment and environmental resources.
Smart Images

Figure CN122450673A_ABST