Robot Motion Outcome Prediction for Fast Collision and Swept Volume Checks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current robotic simulation methods for predicting motion outcomes are time-consuming and require running multiple simulations, especially for collision detection and optimization, which can be inefficient and labor-intensive.
Innovation Solution
A machine learning-based method that uses trained functions to predict motion-outcome data of a robot moving between given locations, eliminating the need for extensive simulations and external modules, allowing for fast and accurate predictions of motion outcomes, including swept volume and collision detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional robotic simulation methods are used to predict motion outcomes, then accuracy can be maintained, but the process becomes extremely time-consuming and computationally intensive
Solution Approach 1:
The system performs preliminary action by pre-computing and storing motion outcome data for numerous robot trajectories in advance. This pre-computed data is then used to quickly answer prediction queries without running full simulations at query time, thus maintaining accuracy while dramatically reducing response time.
Solution Approach 2:
The system creates a copy of the simulation environment and pre-computes motion outcomes in this virtual copy. The results are stored and reused for multiple queries, eliminating the need to run actual simulations repeatedly and reducing both time and computational resource consumption.
2Reliability
If multiple simulations are run for optimization and collision detection, then comprehensive validation is achieved, but the computational resources and time required increase significantly
Solution Approach 1:
The system pre-computes swept volume data and collision information for multiple potential trajectories in advance. When optimization or collision detection is needed, the system queries this pre-computed data rather than running multiple full simulations, maintaining validation reliability while improving productivity.
Solution Approach 2:
The system computes motion outcome data for more trajectories than strictly necessary (excessive action), storing this data for future use. This allows the system to quickly answer multiple queries and perform optimizations without running the full complement of simulations each time, thus improving efficiency while maintaining comprehensive validation capability.
3Adaptability or versatility
If generic robotic motion planner modules are used, then broad applicability is achieved, but accuracy for specific robot configurations is insufficient
Solution Approach 1:
The system applies local quality by customizing the pre-computed motion outcome data for each specific robot configuration, kinematic chain, and tool setup. Instead of using generic data that approximates all robots, the system generates and stores accurate pre-computed data tailored to each specific robot type, maintaining both versatility and precision.
Solution Approach 2:
The system changes parameters by adapting the pre-computation process to specific robot configurations, kinematic parameters, and operational conditions. This allows the same general approach to be applied across different robot types while maintaining accuracy for each specific configuration through parameter-specific pre-computed data.
Data Source
AI summary
Systems and a method for predicting motion-outcome data of a robot moving between a given pair of robotic locations. Data on a given pair of robotic locations are received as input data. A function trained by a machine learning algorithm is applied to the input data, wherein a related robotic motion-outcome data is generated as output data. The robotic motion-outcome data is provided as output data.