LiDAR Object Validation in Tunnels for Stable Vehicle Path Planning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle control systems using LiDAR incorrectly identify tunnel structures or large vehicles as stationary obstacles, leading to erroneous path changes during driving.
Innovation Solution
A vehicle control apparatus that uses LiDAR to divide a specified space into lower and upper spaces, determining the validity of objects within these spaces by analyzing overlap, size, and connectivity, and assigning identifiers to distinguish between different objects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If LiDAR is used to identify external objects in a tunnel, then object detection capability is improved, but false identification of tunnel structures as stationary obstacles occurs
Solution Approach 1:
The patent divides the detection space into multiple regions (upper space and lower space) based on vertical position. By segmenting the space, the system can apply different identification rules to different regions, preventing tunnel ceiling structures in the upper space from being misidentified as stationary obstacles in the lower space where vehicles typically travel.
Solution Approach 2:
The patent introduces a road boundary determination mechanism as an intermediary layer between raw LiDAR detection and obstacle identification. This intermediary step validates detected objects against known road boundaries and spatial constraints, filtering out false positives like tunnel structures before they are classified as obstacles.
2Reliability
If the vehicle control system identifies all detected objects as potential obstacles, then safety is improved, but unnecessary path changes occur due to false obstacle identification
Solution Approach 1:
The patent applies different identification criteria and validation rules to different spatial regions. Objects in the upper space (likely tunnel structures) are evaluated with different standards than objects in the lower space (potential vehicles or obstacles). This local differentiation allows the system to maintain high safety standards while avoiding unnecessary reactions to false obstacles.
Solution Approach 2:
The patent performs preliminary validation of detected objects against road boundaries and spatial constraints before classifying them as obstacles. This preliminary action filters out false positives early in the processing chain, preventing unnecessary path changes while maintaining safety for truly relevant obstacles.
3Area of stationary object
If the LiDAR detects objects in the upper space of a tunnel, then detection coverage is improved, but misidentification of large vehicles or tunnel structures as stationary obstacles occurs
Solution Approach 1:
The patent utilizes the vertical dimension to differentiate between upper space and lower space detections. By adding this dimensional context, the system can distinguish between tunnel ceiling structures (upper space) and vehicles or obstacles (lower space), significantly improving classification accuracy while maintaining comprehensive detection coverage across all vertical levels.
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
Prevents incorrect identification of tunnel structures and large vehicles as stationary obstacles, ensuring accurate navigation and path planning.
Implementation Method 1
identification of an external object using light detection and ranging (LiDAR)
Data Source
AI summary
A vehicle control apparatus and a method thereof are provided. The vehicle control apparatus includes a sensor (e.g., light detection and ranging (LiDAR)), a communication circuit, and a processor. The processor receives a signal indicating that a vehicle has entered a specified space that is identified via a sensor, divides the specified space into a lower space and an upper space based on a vertical axis, generates a lower box corresponding to a first external object in the lower space, generates an upper box corresponding to a second external object in the upper space, determines whether the upper box at least partially overlaps with the lower box, and outputs a signal indicating that the second external object and the first external object are a same object.