Autonomous Vehicle Localization Control Under Obstructed Sensor Views
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In autonomous driving systems, obstacles such as tracks or advertising signboards between sensors and solid objects can hinder precise feature point acquisition, leading to errors in self-location estimation and unstable track generation, compromising safety and ride comfort.
Innovation Solution
A vehicle control device with a self-location estimation processing unit, a solid object movement predicting unit, and a driving movement planning unit that estimates self-location, predicts obstructive object movements, and generates driving plans based on future road situations and position errors to ensure precise navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If feature point matching is performed for self-location estimation, then position information can be obtained, but position error increases when obstructive objects block the sensor view
Solution Approach 1:
The system performs preliminary detection of obstructive objects (tracks, advertising signboards) that may block sensor views of feature points. By identifying these obstacles in advance, the system can predict potential matching failures and take preventive measures such as adjusting sensor parameters or selecting alternative feature points before the matching error occurs.
Solution Approach 2:
The system introduces an intermediary detection and prediction mechanism between the sensor and the feature point matching process. This intermediary layer identifies obstructive objects and predicts their impact on matching accuracy, allowing the system to compensate for potential errors by selecting alternative feature points or adjusting matching parameters.
2Ease of operation
If conventional autonomous driving control is used, then track generation and trace control can be performed, but safety and ride comfort cannot be assured when position error is large
Solution Approach 1:
The system implements feedback by continuously monitoring the presence of obstructive objects and their impact on feature point matching. When obstacles are detected that may cause matching errors, the system adjusts the autonomous driving control in real-time by selecting alternative feature points or modifying track generation parameters, thereby maintaining safety and ride comfort despite the presence of obstacles.
Data Source
AI summary
The present invention provides a vehicle control device capable of ensuring safety and ride comfort. The device includes: a self-location estimation processing unit 306 estimating self location (absolute position information on a map) of a vehicle by matching sensor information obtained from external sensors 2 to 5 mounted in the vehicle with map information including future information as a point group; a solid object movement predicting unit 307 predicting movement of an object as a factor of obstructing the matching with the map information; and a driving movement candidate generating unit (driving movement planning unit) 309 making a driving movement plan of the vehicle on the basis of a result of predicting movement of the obstructive factor object, a result of presuming a road situation at future time, and an estimation result of a position error of the vehicle at future time.