Robot Base Position Planning Using Height Maps and ML Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for determining the position of a robotic device's base to reach interaction points in a physical environment are computationally expensive and time-consuming, leading to inefficient operations and potential hesitation in task execution.
Innovation Solution
A pre-trained machine learning model, such as an artificial neural network, is used to predict candidate positions within the environment based on a height map representation, allowing the robot to plan collision-free trajectories and efficiently reach interaction points.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional methods are used to determine base position, then complete validation of collision-free trajectories is achieved, but the process is computationally expensive and time-consuming
Solution Approach 1:
The system performs preliminary validation of candidate positions using a pre-trained model before executing full trajectory planning. The model predicts valid base positions in advance, filtering out invalid positions before computational resources are spent on detailed trajectory validation, thus reducing overall computation time while maintaining reliability
Solution Approach 2:
The patent replaces conventional computational geometry methods with a machine learning-based pre-trained model. This substitution transforms the mechanical computation process into an intelligent prediction system that can rapidly estimate valid base positions without exhaustive collision checking, significantly reducing computation time while preserving validation accuracy
2Measurement precision
If exhaustive validation of all candidate positions is performed, then complete set of valid positions is obtained, but operational efficiency is reduced
Solution Approach 1:
The pre-trained model performs preliminary filtering of candidate positions before full validation. By predicting which positions are likely to be valid based on learned patterns from training data, the system avoids exhaustive validation of all possible positions, thereby maintaining measurement precision for critical positions while improving overall operational speed
Solution Approach 2:
The system performs partial validation using the pre-trained model's predictions rather than exhaustive validation of all candidate positions. This partial action approach focuses computational resources on the most promising candidate positions identified by the model, achieving sufficient validation accuracy without the time cost of complete validation
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method includes receiving sensor data representative of surfaces in a physical environment containing an interaction point for a robotic device and determining, based on the sensor data, a height map of the surfaces in the physical environment. The method also includes determining, by inputting the height map and the interaction point into a pre-trained model, one or more candidate positions for a base of the robotic device to allow a manipulator of the robotic device to reach the interaction point. The method additionally includes determining a collision-free trajectory to be followed by the manipulator of the robotic device to reach the interaction point when the base of the robotic device is positioned at a selected candidate position of the one or more candidate positions and, based on determining the collision-free trajectory, causing the base of the robotic device to move to the selected candidate position within the physical environment.