LiDAR Material Classification for Accurate Robot Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Robots face challenges in accurately mapping spaces with mixed materials like glass and concrete, as glass can inhibit accurate distance measurement, leading to errors in distinguishing between walls and objects made of different materials.
Innovation Solution
A robot equipped with a LiDAR sensor that uses signal intensity and distance analysis to differentiate between materials like glass and concrete, storing feature information in a map storage unit to create an accurate map, and adjust its operations based on the material features encountered.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a robot uses a LiDAR sensor to sense objects in a space with mixed materials (glass and concrete), then the robot can detect objects and generate a map, but the glass material inhibits accurate distance measurement leading to errors in distinguishing between walls and objects
Solution Approach 1:
The patent segments the sensing process into multiple stages: initial object detection using LiDAR, material classification based on reflection intensity characteristics, and separate handling of glass versus non-glass objects. The controller divides the sensing data into different categories (glass objects, non-glass objects, walls) and processes each category with appropriate algorithms, thereby resolving the contradiction between distance measurement and material distinction accuracy.
Solution Approach 2:
The patent changes the parameter used for object classification from purely distance-based to intensity-based reflection characteristics. By analyzing the reflection intensity pattern (including multiple reflections and transmission characteristics), the system can distinguish glass from non-glass materials even when distance measurements are inaccurate. This parameter change enables reliable material distinction despite the limitations of LiDAR in glass environments.
2Productivity
If the robot senses all objects without material differentiation, then the mapping process is simple and fast, but the robot cannot distinguish between walls and movable objects made of different materials
Solution Approach 1:
The patent performs preliminary material classification during the sensing phase by analyzing reflection intensity characteristics before final map generation. The controller identifies glass versus non-glass objects in real-time during the mapping process, allowing the system to maintain high productivity while preserving material feature information. This preliminary action enables subsequent operations to use the material classification data without requiring additional sensing passes.
Solution Approach 2:
The patent introduces an intermediary classification layer between raw LiDAR sensing and final map generation. The controller acts as an intermediary that processes sensing data through material classification algorithms, extracting material features (glass, non-glass, wall) and passing this enriched information to the map generation module. This intermediary layer preserves material information while maintaining efficient map generation throughput.
3Reliability
If the robot uses signal intensity analysis to distinguish materials, then the robot can differentiate between glass and concrete, but the system complexity increases due to additional analysis requirements
Solution Approach 1:
The patent implements self-service through automated material classification algorithms that run on the robot's existing controller without requiring external assistance or complex additional hardware. The system uses the LiDAR sensor's inherent intensity measurement capability and applies processing algorithms to automatically distinguish materials. This self-service approach achieves reliable material distinction while avoiding the complexity of adding separate specialized sensors or manual classification systems.
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
Enables the robot to generate and update maps accurately, distinguishing between various materials and objects, preventing errors in distance measurement and allowing for safe navigation and operation based on material features.
Implementation Method 1
a light detection and ranging (LiDAR) sensor configured to sense one or more objects provided outside the robot
Data Source
AI summary
Disclosed are a method for drawing a map to which a feature of an object is applied and a robot implementing the same. The robot drawing a map to which feature of an object is applied, which comprises a moving unit configured to control a movement of the robot; a map storage unit configured to store the map to be referred while the robot moves; a sensing unit configured to sense one or more objects provided outside the robot; and a controller configured to control the moving unit, the map storage unit, and the sensing unit, and calculate position information and feature information on the one or more sensed objects, wherein the controller of the robot stores the position information and the feature information of the one or more sensed objects in the map storage unit.