Lane Change Tendency Detection Using LiDAR-Vision Data Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing self-driving technologies face inaccuracies in recognizing vehicle lane change trends due to factors like exhaust gas and dust affecting laser radar technology, and the absence of lane lines in intersection areas hindering machine vision technology.
Innovation Solution
A method and apparatus that combine laser radar and machine vision technologies by obtaining laser point cloud data and scene images, calculating distance relationship values, and using confidence calculations and fusion techniques to comprehensively determine lane change trends, thereby overcoming the limitations of each technology alone.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If lane changing tendency identification is not performed, then the vehicle control system operates normally, but the risk of collision with obstacles in adjacent lanes increases
Solution Approach 1:
The system performs preliminary identification of lane changing tendency by analyzing current steering angle and acceleration data before actual lane changing occurs. This allows the vehicle control system to predict potential lane changes and take preventive measures, reducing collision risk while maintaining manageable system complexity through early detection
Solution Approach 2:
The lane changing tendency identification module acts as an intermediary between the driver's steering input and the vehicle control system. It processes steering angle and acceleration data to generate tendency information, which then informs obstacle detection and avoidance strategies without requiring direct complex control interventions
2Reliability
If comprehensive obstacle detection is performed in adjacent lanes, then collision risk is reduced, but the time and computational resources required increase
Solution Approach 1:
The system focuses obstacle detection resources on specific regions based on lane changing tendency. When tendency is detected, the system intensifies monitoring in the adjacent lane direction; when no tendency is detected, monitoring is reduced or redirected. This localized approach improves detection accuracy for relevant obstacles while reducing overall time and computational resource requirements
Solution Approach 2:
The system dynamically adjusts detection parameters such as scanning frequency, detection range, and sensor activation based on the identified lane changing tendency. High tendency scenarios trigger enhanced detection modes with higher resource consumption, while low tendency scenarios use reduced monitoring modes, optimizing the balance between detection accuracy and resource usage
3Measurement precision
If steering angle and acceleration data are continuously monitored, then lane changing tendency identification accuracy is improved, but energy consumption and processing load increase
Solution Approach 1:
The system implements periodic sampling of steering angle and acceleration data rather than continuous monitoring. By selecting appropriate sampling intervals, the system maintains sufficient accuracy for tendency identification while significantly reducing the energy consumption and processing load associated with continuous data acquisition and analysis
Data Source
Figure 1
Figure 2
Figure 3
AI summary
This application provides a method and an apparatus for recognizing a vehicle lane change trend, and belongs to the field of self-driving technologies. This application may be applied to an intelligent vehicle, a new energy vehicle, or a connected vehicle. The method for recognizing the vehicle lane change trend includes: obtaining laser point cloud data of a detected target vehicle (201); obtaining, based on the laser point cloud data, a first distance relationship value between a center line of a lane in which a current vehicle is located and the target vehicle (202); obtaining a scene image including the target vehicle (203); obtaining a second distance relationship value between the center line and the target vehicle based on the scene image (204); calculating first confidence of a plurality of obtained first distance relationship values and second confidence of a plurality of obtained second distance relationship values (205); calculating a plurality of fusion distance relationship values of the plurality of first distance relationship values and the plurality of second distance relationship values based on the first confidence and the second confidence (206); and determining, based on the plurality of fusion distance relationship values, whether the target vehicle has a lane change trend (207). By using this method and apparatus, the vehicle lane change trend can be more accurately recognized.