Hybrid Robot Path Planning for Congested Environments
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robot path planning technologies fail to efficiently navigate robots in dynamic environments with human and robot congestion, leading to sub-optimal plans and scalability issues.
Innovation Solution
A hybrid robot path planning approach that integrates hierarchical planning with low-level motion planning and a controller, using concepts like critical regions and region-based Voronoi diagrams to construct a discrete abstract model, and a custom cost function to account for human and robot congestion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional path planning methods are used in dynamic environments with humans and robots, then the planning process becomes computationally expensive and slow, but the path safety and efficiency deteriorate due to inability to handle congestion
Solution Approach 1:
The patent segments the path planning problem into two distinct levels: high-level abstract path planning that determines the sequence of regions to visit, and low-level motion planning that generates specific trajectories within each region. This segmentation allows the high-level planner to focus on safety and congestion avoidance using abstract state representations, while the low-level planner handles detailed motion execution, thereby improving both path safety and reducing planning time.
Solution Approach 2:
The patent introduces an abstract state representation as an intermediary layer between the physical environment and the path planning algorithm. This abstraction maps continuous physical spaces to discrete states, enabling efficient computation of safe paths while accounting for human and robot congestion without requiring exhaustive simulation of all possible motions in the continuous space.
2Reliability
If centralized control is used to coordinate multiple robots, then collision avoidance improves, but the system complexity and computational burden increase significantly
Solution Approach 1:
The patent divides multi-robot path planning into independent high-level abstract path planning for each robot, where each robot computes its own safe path based on abstract state transitions. This segmentation reduces control system complexity by eliminating the need for centralized coordination while maintaining collision avoidance through the abstract state framework that inherently accounts for other robots' positions and movements.
3Manufacturing precision
If detailed motion planning is performed in continuous space, then path accuracy improves, but the computational cost and planning time increase exponentially
Solution Approach 1:
The patent segments the planning process into high-level abstract path planning in discrete state space and low-level motion planning in continuous space. The high-level planner quickly determines the sequence of regions to visit using abstract state representations, providing a coarse but safe path. The low-level planner then refines this into an accurate motion trajectory within each region, achieving path accuracy while keeping total planning time manageable through the two-stage approach.
Solution Approach 2:
The patent performs preliminary abstract path planning to determine the sequence of regions and identify safe paths before executing detailed motion planning. This preliminary action establishes a framework that guides the subsequent low-level motion planning, reducing the search space and computational burden while ensuring path safety and accuracy.
Data Source
AI summary
A computing system may include a processor. The computing system may include a memory having a set of instructions, which when executed by the processor, cause the computing system to determine paths in an abstract state for a first robot to progress from an origin to a destination, determine a cost for each respective path of the paths based on a congestion measurement of the respective path and a measurement of the abstract state, select a selected path from the paths that is associated with a lowest cost from the costs, and compute a motion plan for the first robot to traverse the selected path.