Robot Ground Segmentation and Object Classification for Path Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robot technologies face challenges in accurately identifying external objects and planning movement paths due to limitations in sensor data processing and map integration, leading to inefficiencies in navigation and control.
Innovation Solution
A robot control apparatus and method utilizing LiDAR and grid maps to segment sensor data into areas, determine ground vs. non-ground regions, and classify objects as static, dynamic, or mixed, enabling precise robot movement control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sensor data is processed using traditional methods without segmentation, then the processing is simpler, but the identification accuracy of external objects deteriorates
Solution Approach 1:
The patent divides the sensor data processing into segmented areas based on distance ranges. The processing space is divided into multiple regions (e.g., first area, second area, third area) with different thresholds and processing methods for each segment, allowing accurate identification while managing computational complexity through localized processing rules.
2Measurement precision
If comprehensive sensor data processing is performed to accurately identify all objects, then the identification accuracy improves, but the computational load increases
Solution Approach 1:
The patent segments the processing based on distance areas, applying different computational thresholds to different regions. Close-range areas use stricter thresholds while far-range areas use more lenient ones, reducing overall computational load while maintaining accuracy where it matters most.
Solution Approach 2:
Different processing qualities and thresholds are applied to different spatial regions. The first area (closest to robot) uses the most stringent processing, the second area uses intermediate processing, and the third area uses the least processing, optimizing energy usage based on local requirements.
3Reliability
If detailed object classification is performed to distinguish static, dynamic, and mixed objects, then the navigation safety improves, but the processing time increases
Solution Approach 1:
The patent classifies objects into different types (static, dynamic, mixed) based on their movement characteristics detected across segmented areas. By processing objects in distance-based segments and applying type-specific navigation rules, the system ensures safety while reducing processing time through targeted classification rather than exhaustive analysis.
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 accurate and efficient identification of external objects with reduced computational load, allowing for safer and more controlled robot navigation by distinguishing between different types of objects and generating optimal movement paths.
Implementation Method 1
A robot control apparatus and method utilizing LiDAR and grid maps to segment sensor data into areas
Data Source
AI summary
A robot control apparatus and a method thereof are provided. The robot control apparatus includes at least one sensor and a processor. The processor may be configured to obtain, via the at least one sensor and from an area within a designated distance from a robot, data points associated with one or more target objects around the robot, determine a first group of data points and a second group of data points, determine, based on an angle difference between the first group and the second group and based on a height difference between the first group and the second group, whether the first segmented area corresponds to a ground, and control, based on the determination of whether the first segmented area corresponds to the ground, movement of the robot.