Robot Path Generation Using Voxelized Scene and Necessary Points

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The heavy workload associated with setting paths for robots in different scenarios necessitates an efficient path generation method to reduce the burden and improve flexibility.

Innovation Solution

A path generation method for robots that involves scene voxelization, machine modeling, determining initial points, performing collision and stability detection, removing redundant points, and connecting necessary points to generate efficient paths autonomously.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If different paths are set for each robot in combination with different scenarios, then the robot can adapt to various scenarios, but the workload of setting the path becomes relatively heavy

Engineering Contradiction:
Improveadaptability to different scenariosVSAvoidworkload of setting path
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The robot autonomously generates its own path by performing voxelization of the current scene, establishing a scene model, determining initial points, and executing collision detection and stability detection without requiring external path setting for each scenario

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The robot performs scene voxelization and establishes the scene model in advance before path generation, preparing the environment data structure beforehand to enable efficient autonomous path planning when needed

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS12461530B2Path generation method for robot, robot and computer-readable storage medium
Publication Date: 2025.11.04 UBTECH ROBOTICS CORP LTD
  • US12461530B2 patent drawing
  • US12461530B2 patent drawing
  • US12461530B2 patent drawing

AI summary

A path generation method for a robot includes: based on a scene model of a current scene and a machine model corresponding to the robot, combined with multiple configurations of the robot, determining a plurality of initial points between a start point and an end point in the scene model; removing redundant points from the plurality of the initial points to obtain necessary points among the initial points; and connecting the necessary points to obtain a path between the start point and the end point.