Moving Robot Topological Mapping for Accurate Node Connections
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing moving robots face challenges in accurately generating maps and recognizing positions due to inaccurate connection relationships between nodes, difficulties in distinguishing dynamic and fixed obstacles, and acquiring information on intersecting paths, leading to inefficient navigation.
Innovation Solution
The implementation of a moving robot equipped with a lidar sensor, cameras, and advanced sensors for terrain and obstacle detection, combined with a controller that generates and updates a topological map in real-time, determines open movement directions, and adjusts its path to minimize obstacle avoidance actions, allowing for more accurate node connection and stable navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a distance sensor is used to generate an initial map and a global topological map is generated by setting nodes through pixel processing, then a map can be generated, but the connecting relationship between nodes becomes inaccurate
Solution Approach 1:
The patent transitions from 2D pixel-based map processing to 3D spatial coordinate-based map generation. By using depth information from the distance sensor to create three-dimensional coordinates for nodes, the system preserves spatial relationships and improves connection accuracy between nodes in the topological map.
Solution Approach 2:
The patent introduces an intermediary processing step that converts distance sensor data into structured spatial coordinates before generating the topological map. This intermediary coordinate system serves as a bridge between raw sensor data and the final map structure, enabling accurate node connection relationships.
2Reliability
If distance information is acquired while moving to generate a map, then the map can be created, but it becomes difficult to distinguish dynamic obstacles from fixed obstacles
Solution Approach 1:
The patent performs preliminary classification of obstacles by analyzing the temporal changes in distance measurements. By comparing distance data across multiple time points and positions, the system pre-identifies dynamic versus fixed obstacles before final map generation, improving classification reliability without requiring additional sensors.
Solution Approach 2:
The patent implements a feedback mechanism where the robot continuously updates its understanding of the environment by comparing new sensor readings with existing map data. This feedback loop enables the system to distinguish dynamic obstacles from fixed ones by detecting changes in obstacle positions and characteristics over time.
3Loss of information
If nodes are set based on pixel processing of an initial map, then a topological map can be generated, but information on points where moving paths intersect is lost
Solution Approach 1:
The patent recovers path intersection information by utilizing the third dimension (depth) in coordinate processing. By analyzing spatial coordinates in three dimensions rather than just 2D pixel positions, the system can identify points where paths intersect and preserve this critical navigation information in the topological map.
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
This approach enables the moving robot to generate accurate topological maps, improve navigation stability, and enhance obstacle avoidance, resulting in more efficient and precise movement within its environment.
Implementation Method 1
a lidar sensor that acquires terrain information on a traveling area
Implementation Method 2
a lidar sensor that acquires terrain information on a traveling area by using a laser
Data Source
AI summary
A moving robot includes: a sensor to acquire terrain information; a memory to store node data for at least one node; and a controller, and the controller determines whether at least one open movement direction exists among a plurality of movement directions, based on sensing data and the node data, generates a new node in the node data when at least one open movement direction exists, determines any one of the open movement directions as a traveling direction for the robot, determines whether at least one of the nodes needs to be updated exists, based on the node data when the open movement direction does not exist, controls the moving robot to move to one of the nodes that need to be updated, and generates of a map including the at least one node, based on the node data, when the node that needs to be updated does not exist.