Vehicle Position Estimation Using Weighted Multi-Sensor Feature Points
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing position estimation systems for vehicles suffer from reduced accuracy due to relaxed matching conditions between in-vehicle camera images and reference data, particularly in challenging environments, leading to inaccurate vehicle positioning.
Innovation Solution
A position estimation system utilizing a front sensor and peripheral sensors to detect feature points, weighing information based on vehicle state and environment, and selectively using feature point groups for accurate positioning through a control device.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the matching condition between scene image and reference data is relaxed to increase matching opportunities, then the frequency of position determination is improved, but the accuracy of position determination deteriorates
Solution Approach 1:
The patent applies dynamics by making the determination threshold variable rather than fixed. The threshold dynamically adjusts based on vehicle speed and steering amount - decreasing as speed increases or steering amount increases, and increasing otherwise. This dynamic adjustment allows the system to maintain appropriate matching stringency across different operating conditions, resolving the contradiction between matching frequency and accuracy.
Solution Approach 2:
The patent changes the parameter of determination threshold based on vehicle state parameters (speed and steering amount). By modifying the threshold parameter according to these conditions, the system optimizes the balance between detecting enough features for positioning and maintaining accuracy, thus resolving the contradiction between productivity and measurement precision.
2Measurement precision
If the determination threshold is increased to maintain accuracy, then the accuracy of position determination is improved, but the opportunities for matching decrease
Solution Approach 1:
The determination threshold dynamically adapts to vehicle conditions - being lowered when vehicle speed is high or steering amount is large (conditions where fewer stable features are available), and raised when conditions are favorable. This dynamic behavior maintains accuracy while ensuring positioning can still occur in challenging conditions.
Solution Approach 2:
The system modifies the determination threshold parameter based on measured vehicle state parameters (speed and steering amount), allowing the threshold to change adaptively rather than remaining fixed, thus balancing accuracy requirements with the need to maintain positioning frequency.
3Measurement precision
If multiple sensors are used to detect feature points, then the accuracy of position estimation is improved, but the device complexity increases
Solution Approach 1:
The patent merges data from multiple sensor types (laser radar, ultrasonic sensor, imaging sensor) into a unified feature detection system. By combining these sensors and their detection results, the system achieves improved position estimation accuracy while managing complexity through integrated processing rather than separate independent systems.
Solution Approach 2:
The system employs multi-functional sensors that can detect various types of features (distance, acoustic signals, visual images) and processes them through a common feature point detection framework. This universal approach allows one system to serve multiple detection purposes, improving accuracy without proportionally increasing complexity.
Data Source
AI summary
A position estimation system provided in a vehicle can include a front sensor, a peripheral sensor, a position estimation device, and a control device. The position estimation device can be configured to determine weights for each of a plurality of pieces of information related to a state and a surrounding environment of the vehicle, the plurality of pieces of information comprising vehicle speed, steering angle, traveling environment, predicted curvature, and surrounding environment. The position estimation device can then select whether to use a plurality of feature point groups including a part of the plurality of feature points detected by the front sensor and the peripheral sensor to estimate the position of the vehicle based on the weights.