Robot Laser Guidance for Narrow-Passage Collision Avoidance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robot path planning algorithms are prone to deviations in positioning coordinates, especially when navigating narrow passages, increasing the likelihood of collisions with obstacles.
Innovation Solution
Construct a closed plane graph based on the robot's chassis size, perform laser irradiation to acquire a laser point set, and control the robot's movement state based on the farthest distance between the center point and the laser points to avoid collisions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If global path planning algorithm and local path planning algorithm are used to determine robot movement path, then the robot can navigate from starting point to end point, but positioning coordinate deviations occur especially in narrow passages increasing collision probability
Solution Approach 1:
The patent segments the laser detection area into multiple regions by constructing a closed plane graph with multiple edges radiating from the center point. Each edge corresponds to a specific angular sector, allowing independent detection and analysis of obstacles in different directions. This segmentation enables more precise positioning by identifying which specific segment contains the obstacle, thereby reducing coordinate deviation in narrow passages.
Solution Approach 2:
The patent introduces an angular dimension by constructing a closed plane graph with multiple edges at different angles. Instead of only radial distance measurement, the system now considers both radial distance and angular position. This dimensional expansion allows the robot to determine its position more accurately by combining distance information from multiple angular perspectives, resolving the positioning accuracy issue in narrow passages.
2Reliability
If laser irradiation is performed to detect obstacles, then obstacle detection capability is improved, but false detection may occur leading to incorrect movement control
Solution Approach 1:
The patent applies local quality by assigning different weights or priorities to different edges of the closed plane graph based on their angular positions and distances. Edges that are more likely to detect true obstacles (e.g., those pointing toward known navigation paths) are given higher reliability weights, while edges prone to false detection (e.g., those pointing toward walls or fixed structures) are given lower weights. This localized quality adjustment improves overall detection accuracy without requiring complete system redesign.
Solution Approach 2:
The patent implements feedback by using the detected obstacle information to adjust and optimize the closed plane graph construction in subsequent detection cycles. The system learns from previous detection results, identifying which edges produce reliable detections and which produce false alarms. This feedback mechanism continuously refines the detection strategy, improving reliability while maintaining manageable system complexity through adaptive rather than static configurations.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Reduces the probability of collisions when entering narrow passages by accurately guiding the robot's movement, enhancing its security during navigation.
Implementation Method 1
a laser transmitter configured to perform laser irradiation to acquire coordinates of laser points formed by the laser irradiation on the obstacle
Data Source
AI summary
This disclosure provides a method and apparatus for controlling a robot, and a non-transitory computer-readable storage medium, and relates to the technical field of robot. The method of controlling a robot therein includes: constructing a closed plane graph according to a size of a chassis of the robot, the closed plane graph passing through a center point of the robot chassis and a target point on a planning path of the robot, a connection line between the center point and the target point being a symmetry axis of the closed plane graph; performing laser irradiation from the center point to an area of the closed plane graph to acquire a laser point set; and controlling a movement state of the robot according to a farthest distance between all the laser points in the laser point set and the center point.