Robot Path Planning Using Neural Networks for Interference Avoidance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robot path generation methods rely heavily on manual teaching or random sampling, leading to variability and suboptimal paths in terms of evaluation criteria such as electric power consumption, and lack efficiency in avoiding interference with workpieces.
Innovation Solution
A robot path generation device employing a track planning module that uses machine learning based on a dataset of path data and evaluation values to generate paths between set start and end points, ensuring the robot avoids interference through a neural network model that learns to optimize path quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If manual teaching or random sampling is used for robot path generation, then the path can be generated without complex learning systems, but the path quality and consistency are suboptimal and variable
Solution Approach 1:
The system performs preliminary machine learning training offline to build a neural network model that stores optimal path planning knowledge. This pre-computed model is then reused during actual robot operations, avoiding the need for complex real-time learning while achieving high path quality and consistency.
2Reliability
If traditional path generation methods are used, then the system structure remains simple, but the ability to avoid interference with workpieces is insufficient
Solution Approach 1:
The patent replaces traditional mechanical or rule-based path generation methods with a data-driven neural network model. The neural network learns from training data to predict optimal paths that avoid workpiece interference, substituting complex mechanical judgment systems with a learned mathematical model.
3Use of energy by moving object
If evaluation criteria such as electric power consumption are not optimized, then the path generation process is simpler, but the energy efficiency and performance are suboptimal
Solution Approach 1:
The system incorporates multiple evaluation criteria including electric power consumption, path length, and interference avoidance as optimization parameters in the neural network training process. By adjusting these parameters during training, the model learns to generate paths that optimize energy efficiency while maintaining other performance requirements.
Data Source
AI summary
To generate a more appropriate path, provided is a robot path generation device including circuitry configured to: hold a track planning module learning data set, in which a plurality of pieces of path data generated based on a motion constraint condition of a robot, and evaluation value data, which corresponds to each of the plurality of pieces path data and is a measure under a predetermined evaluation criterion, are associated with each other; and generate, based on a result of a machine learning process that is based on the track planning module learning data set, a path of the robot between a set start point and a set end point, which are freely set.