Vehicle Position Estimation by Lane and Odometry Fusion at Low Speed
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In automated driving systems, positional deviations in the width direction of a road can occur due to incorrect recognition of lane division lines, leading to unnecessary operations or control interventions, especially in low-speed ranges where reliance on single-sided lane recognition is common.
Innovation Solution
A position determination device that combines camera-based lane recognition with vehicle movement data using a complementary filter, weighting the movement-based position estimation more heavily at low speeds to improve accuracy and stability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If position is determined using only division line recognition, then device complexity is reduced, but measurement precision deteriorates in low-speed ranges
Solution Approach 1:
The patent combines two position determination methods: division line recognition and odometry-based position estimation. The position determination unit synthesizes both positions to determine the final vehicle position, merging the advantages of both methods while mitigating their individual weaknesses, especially in low-speed ranges where odometry provides superior accuracy
Solution Approach 2:
The position determination unit dynamically adjusts the weighting between division line position and odometry position based on vehicle speed. At low speeds, it prioritizes odometry-based position estimation, while at higher speeds, it relies more on division line recognition, making the system adaptive to different operating conditions
2Ease of operation
If division line recognition is used for position determination, then ease of operation is improved, but reliability deteriorates due to positional deviation
Solution Approach 1:
The system uses feedback from multiple sources (division line recognition and odometry) to continuously refine position determination. The position determination unit compares and synthesizes positions from both methods, using the feedback loop to correct deviations and improve reliability, especially when division line recognition alone would cause errors
Solution Approach 2:
The position determination unit acts as an intermediary that reconciles the outputs from division line recognition and odometry-based estimation. It mediates between these two position sources, synthesizing them into a reliable final position determination that reduces the impact of recognition errors
3Ease of operation
If camera-based lane recognition is used, then ease of operation is improved, but measurement precision worsens due to recognition errors of deceleration displays
Solution Approach 1:
The system extracts position information from multiple sources rather than relying solely on camera-based lane recognition. By taking out the odometry-based position estimation as a separate, independent position source, the system avoids the pitfalls of misrecognizing deceleration display markings as actual lane division lines
Data Source
AI summary
A position determination device comprises a camera to capture an image ahead of a vehicle. The position determination device acquires position of the vehicle; estimates a first position of the vehicle, based on a position of a lane included in the image, the position of the vehicle that has been acquired, and map information; estimates a second position of the vehicle, based on a position of the vehicle that was determined most recently and a movement amount of the vehicle; and synthesizes the first position and the second position to determine the position of the vehicle, wherein the device weights the second position while the vehicle is traveling at a speed equal to or lower than a predetermined speed, and synthesizes the first position and the second position.