Robot Map Building via Dynamic Obstacle Removal
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional SLAM methods fail to accurately distinguish dynamic obstacles from static objects, leading to errors in map data and potential collisions, as they can only approximate the robot's position and recognize obstacles as either static or dynamic.
Innovation Solution
A method using three-dimensional surface data matching, specifically through iterative closest point (ICP) algorithms, to detect and remove dynamic obstacles by calculating distance differences and converting data into two-dimensional grid images for error correction, ensuring accurate map building based on static objects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional SLAM method is used to build map, then map building can be performed, but dynamic obstacles cannot be accurately distinguished from static objects leading to map data errors
Solution Approach 1:
The patent applies dynamics principle by capturing multiple surface data sets at different time points and comparing them to detect changes. By sequentially acquiring first and second surface data and calculating differences between them, the system dynamically identifies moving objects that change position between captures, thereby distinguishing dynamic obstacles from static objects with high accuracy.
Solution Approach 2:
The patent implements feedback by using the detected dynamic obstacle information to correct the map building process. The map generation unit receives feedback about which objects are dynamic versus static, and accordingly generates accurate map data containing only static objects, preventing map distortion and ensuring reliable navigation data.
2Productivity
If dynamic obstacle is recognized as static object, then map data can be generated, but map data becomes distorted and robot position includes errors
Solution Approach 1:
The patent applies preliminary action by detecting and removing dynamic obstacles from surface data before the map generation process. The moving object detection unit identifies dynamic objects in advance by comparing sequential surface data, and the map generation unit excludes these detected dynamic objects when creating the map, ensuring position accuracy from the outset rather than correcting errors later.
3Reliability
If dynamic obstacle is recognized as static object, then map building proceeds, but robot cannot avoid dynamic obstacle causing collision
Solution Approach 1:
The patent applies segmentation by separating the obstacle detection and classification function from the general map building process. The moving object detection unit is a dedicated component that specifically analyzes surface data changes to identify dynamic obstacles, while the map generation unit handles static object mapping. This segmentation allows the system to maintain simple, reliable collision avoidance for dynamic objects while keeping the overall system architecture modular and manageable.
Data Source
AI summary
Disclosed are a robot, which builds a map using a surface data of a three-dimensional image, from which a dynamic obstacle is removed, and a method of building a map for the robot. The method includes sequentially acquiring first and second surface data of a route on which the robot moves; matching the first and second surface data with each other to calculate a difference between the first and second surface data; detecting a dynamic obstacle from the first and second surface data according to the difference between the first and second surface data; generating a third surface data by removing the dynamic obstacle from at least one of the first and second surface data; and matching the third surface data and any one of the first and second surface data with each other to build the map.