Autonomous Driving Position Detection via Scene-Adaptive Switching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous driving systems face challenges in accurately detecting vehicle position, particularly due to limitations in hybrid navigation systems that may not provide sufficient accuracy in various travel situations.
Innovation Solution
An autonomous driving system that includes a positioning unit, a map database, and multiple detection units to identify traveling scenes and detect vehicle positions using camera and radar sensor data, allowing for adaptive position detection processing to enhance accuracy and prevent autonomous driving control based on incorrect positions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If hybrid navigation with map matching processing is used to detect vehicle position, then the system can provide a vehicle position on the map, but the detection accuracy is insufficient in various travel situations
Solution Approach 1:
The system dynamically switches between different position detection methods (hybrid navigation and recognition-based detection) based on the identified traveling scene. The electronic controller selects the appropriate detection method according to the current situation, making the system adaptive to various travel conditions while maintaining high accuracy
Solution Approach 2:
The system changes the detection parameters and methods based on the traveling scene identification. Different detection algorithms and sensor combinations are applied depending on the scene type (e.g., urban, rural, highway), optimizing accuracy for each specific condition
2Measurement precision
If recognition-based position detection is used to improve accuracy in specific scenes, then detection accuracy increases, but the system complexity increases due to multiple detection units and scene identification
Solution Approach 1:
The electronic controller serves multiple functions: it manages the positioning unit, identifies traveling scenes, selects appropriate detection methods, and integrates results from multiple sensors. This multi-functionality reduces the need for separate dedicated control units for each function, thereby managing system complexity while achieving high accuracy through recognition-based detection
3Measurement precision
If the system uses multiple detection methods and scene identification, then position detection accuracy improves, but the processing time and computational load increase
Solution Approach 1:
The system performs traveling scene identification in advance before executing the position detection. By pre-classifying the current scene type, the system can quickly select the appropriate detection method without performing complex analysis during the actual position detection phase, thereby reducing processing time while maintaining high accuracy
Data Source
AI summary
An autonomous driving system includes a positioning unit that measures a position of a vehicle; a database that stores map information; an actuator that controls traveling of the vehicle; and an electronic controller configured to process: a detection of a first vehicle position and the map information; an identification of a traveling scene based on the first vehicle position and the map information; a detection of a second vehicle position by preforming position detection processing; and a control of the actuator based on the second vehicle position if a distance between the first vehicle position and the second vehicle position is equal to or smaller than a threshold or a control of the actuator based on the first vehicle position if the distance is not equal to or smaller than the threshold.