Hierarchical Polytope Mapping for Real-Time HRC Motion Planning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for human-robot collaboration in dynamic environments are computationally expensive and prone to noise, failing to efficiently handle topological queries and collision-free motion planning due to the complexity of computing overlapping polytopes from point clouds.
Innovation Solution
A method using a linear octree structure to iteratively update convex free-space regions around points of interest, leveraging depth cameras for robot removal and voxelized representations, and constructing convex polytopes for efficient motion planning and intent prediction, enabling real-time collision-free motion planning and intent prediction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If overlapping polytopes are computed from point clouds for human-robot collaboration, then motion planning and intent prediction can be performed, but computational time increases and robustness decreases due to noise and complexity
Solution Approach 1:
The patent segments the workspace into a hierarchical structure of convex polytopes organized in an octree, where each node represents a convex region. This segmentation allows the system to process only relevant local regions rather than computing all overlapping polytopes globally, significantly reducing computational time while maintaining motion planning robustness through the hierarchical organization that enables efficient querying at appropriate levels of detail.
Solution Approach 2:
The patent performs preliminary construction of the hierarchical convex polytope structure from the point cloud before motion planning queries are executed. By pre-processing the environment into a structured hierarchical representation, the system avoids repeated expensive computations during real-time operation, reducing computational time for subsequent motion planning while improving robustness through a stable, pre-validated spatial model.
2Productivity
If traditional methods are used for topological mapping and motion planning in dynamic environments, then complete coverage can be achieved, but the system becomes computationally expensive and slow
Solution Approach 1:
The workspace is segmented into a hierarchical octree structure where each node contains a convex polytope representation. This segmentation enables the motion planning algorithm to operate at different levels of the hierarchy, performing quick queries on coarser levels when possible and only refining to finer levels when necessary, thereby increasing productivity while reducing computational energy consumption compared to processing the entire workspace at maximum detail.
Solution Approach 2:
The patent introduces a hierarchical dimension to the traditional topological mapping by organizing convex polytopes in an octree structure with multiple levels of abstraction. This additional hierarchical dimension allows the system to answer topological queries efficiently by operating at the appropriate level of detail, increasing productivity while minimizing computational energy through selective refinement rather than exhaustive processing.
3Speed
If depth cameras are used for robot removal and voxelized representations, then real-time processing is enabled, but noise and measurement errors increase
Solution Approach 1:
The patent segments the noisy point cloud data into discrete convex polytopes at different hierarchical levels. By organizing the data this way, the system can perform robust geometric computations on simplified convex representations rather than processing raw noisy points directly, enabling real-time processing speed while mitigating measurement errors through the noise-filtering property of convex hull computations at each hierarchical level.
Solution Approach 2:
The system performs preliminary voxelization and convex polytope construction from the depth camera data before motion planning queries. This pre-processing step converts noisy continuous point cloud data into a discrete hierarchical structure, enabling real-time processing speed for subsequent queries while improving measurement precision by filtering noise through the convexification process that inherently smooths out minor measurement variations.
Data Source
AI summary
Various aspects of techniques, systems, and use cases for robot movement within human-robot collaboration (HRC) areas are disclosed. Convex free-space regions around points of interest in the HRC are detected and updated using one or more sensors. Collision-free motion plans for robots use a Hierarchical Convex Polytope (HCP) region in which the robot is removed from the occupied space in the scene using depth cameras.