Robot Motion Optimization Using 3D Distance Field Constraints
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robot collision avoidance techniques require tedious and time-consuming setup of geometry primitives, and computation time increases dramatically with the number of obstacles, making real-time collision avoidance impractical.
Innovation Solution
A robot collision avoidance motion optimization technique using a distance field constraint function, where CAD or sensor data is converted to voxels and a 3D binary matrix is created, allowing for rapid computation of a robot path that avoids obstacles without the need for geometry primitives.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If geometry primitives are used to approximate robot and obstacle shapes for collision avoidance, then the complexity of motion optimization calculation is reduced to a manageable level, but the setup process becomes tedious and time-consuming
Solution Approach 1:
The patent replaces the manual mechanical process of defining geometry primitives with an automated computational approach. The system automatically generates collision avoidance constraints by processing CAD models and sensor data through algorithms that compute distance fields and generate constraints without human intervention, eliminating the tedious setup process while maintaining calculation simplicity
Solution Approach 2:
The system performs preliminary processing of obstacle and robot geometry data before motion optimization. By pre-computing distance fields from CAD models and sensor data, and pre-generating collision avoidance constraints based on these distance fields, the system prepares all necessary information in advance, making the actual motion optimization faster and eliminating the need for time-consuming setup during operation
2Reliability
If traditional motion planning optimization is used, then collision avoidance can be achieved, but computation time increases dramatically with the number of obstacles making real-time operation impractical
Solution Approach 1:
The patent segments the collision avoidance problem into two independent parts: (1) pre-computation of distance fields from obstacle geometries, and (2) generation of convex constraints using these distance fields. This segmentation allows the computationally intensive part to be done once offline, while online operation only requires evaluating pre-computed distance fields and solving a convex optimization problem, making computation time independent of the number of obstacles
Solution Approach 2:
The system changes the mathematical formulation from traditional non-convex collision constraints to convex constraints based on distance field parameters. By representing obstacles through distance fields and formulating constraints as convex inequalities involving these fields, the optimization problem becomes computationally tractable with polynomial-time complexity, enabling real-time solution regardless of obstacle count
Data Source
AI summary
A robot collision avoidance motion optimization technique using a distance field constraint function. CAD or sensor data depicting obstacles in a robot workspace are converted to voxels, and a three-dimensional binary matrix of voxel occupancy is created. A corresponding distance map matrix is then computed, where each cell in the distance map matrix contains a distance to a nearest occupied cell. The distance map matrix is used as a constraint function in a motion planning optimization problem, where the optimization problem is convexified and then iteratively solved to yield a robot motion profile which avoids the obstacles and minimizes an objective function such as distance traveled. The distance field optimization technique is quickly computed and has a computation time which is independent of the number of obstacles. The disclosed optimization technique is easy to set up, as it requires no creation of geometry primitives to approximate robot and obstacle shapes.