Robot Motion Planning Graph Switching for Changing Tool Geometry
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current robot motion planning systems face challenges in efficiently planning paths at low costs and fast speeds, especially when the robot's characteristics change, such as when holding objects of varying sizes or switching end effectors, due to limitations in storing and processing motion planning information.
Innovation Solution
The system generates and stores multiple planning graphs and edge information based on different sets of physical dimensions for a robot, allowing dynamic switching between them to accommodate changing robot configurations, and uses discretized representations of the environment to perform collision checking and optimization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple planning graphs are stored for different robot configurations, then adaptability to changing robot characteristics is improved, but memory storage requirements increase
Solution Approach 1:
The system segments the planning information into multiple discrete planning graphs, each corresponding to a specific robot configuration (e.g., different end effectors, different held objects). Instead of storing one large comprehensive planning structure, the system divides it into manageable segments that can be selectively loaded and stored in memory based on current robot characteristics, thus reducing overall storage requirements while maintaining adaptability.
Solution Approach 2:
The system dynamically selects and loads the appropriate planning graph based on real-time robot characteristics such as the current end effector configuration or objects being held. This dynamic approach allows the system to have multiple planning graphs available but only load the necessary ones into memory at any given time, balancing adaptability with memory constraints.
2Speed
If motion planning information is pre-computed and stored, then processing speed is improved, but the amount of information that can be stored on processor chip circuitry is limited
Solution Approach 1:
The system extracts only the essential planning graph structures and edge information needed for motion planning from comprehensive pre-computed data. By taking out only the critical components (nodes representing robot states and edges representing transitions) rather than storing complete motion planning databases, the system achieves fast processing with reduced memory footprint suitable for processor chip circuitry.
Solution Approach 2:
The planning graphs are pre-computed offline for various robot configurations, but only the essential structural information is stored on the processor chip. This preliminary action allows the system to have ready-to-use planning structures during operation without requiring extensive on-chip storage, as the detailed computations were performed in advance.
3Adaptability or versatility
If the robot changes physical dimensions (e.g., holding objects, switching end effectors), then operational versatility is improved, but collision detection complexity increases
Solution Approach 1:
The system applies local quality by creating specific planning graphs tailored to each robot configuration's local characteristics. When the robot holds different objects or uses different end effectors, the appropriate pre-computed planning graph with its specific collision information is selected, rather than using a generic complex collision detection algorithm for all scenarios. This reduces the effective complexity at runtime.
Solution Approach 2:
The system manages complexity by changing parameters - specifically, by switching between different pre-computed planning graphs that have different physical dimension parameters built in. Each planning graph contains collision information optimized for its specific robot configuration, allowing the system to handle various operational scenarios without requiring a single overly complex collision detection system.
Data Source
AI summary
A robot control system determines which of a number of discretizations to use to generate discretized representations of robot swept volumes and to generate discretized representations of the environment in which the robot will operate. Obstacle voxels (or boxes) representing the environment and obstacles therein are streamed into the processor and stored in on-chip environment memory. At runtime, the robot control system may dynamically switch between multiple motion planning graphs stored in off-chip or on-chip memory. The dynamically switching between multiple motion planning graphs at runtime enables the robot to perform motion planning at a relatively low cost as characteristics of the robot itself change.