Methods and systems for vehicle control under degraded lane perception range
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous vehicle systems struggle to maintain effective lateral and longitudinal control under degraded perception conditions, such as fog, rain, or imperceptible lane markings, which can lead to the need for human intervention.
Innovation Solution
A vehicle control system that includes a sensing system, a perception system, an Electronic Power Steering (EPS) system, and a processor configured to execute program instructions for implementing a motion planner and a lateral controller in a control loop. The system switches between modes based on the availability of lane information and look-ahead points to maintain vehicle trajectory and reduce the need for human intervention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the look-ahead point is set far from the vehicle to improve trajectory planning accuracy, then the desired trajectory can be determined more accurately, but the look-ahead point may extend beyond the perception range under degraded conditions
Solution Approach 1:
The system dynamically adjusts the look-ahead point distance based on perception range conditions. When perception is degraded, the look-ahead point is moved closer to the vehicle; when perception is good, it is positioned farther away. This dynamic adaptation resolves the contradiction between needing far look-ahead points for accuracy and near look-ahead points for reliability under degraded conditions.
Solution Approach 2:
The system changes the parameter of look-ahead point distance based on perception quality. By monitoring perception range and adjusting the look-ahead distance parameter accordingly, the system maintains both trajectory planning accuracy and perception coverage across varying environmental conditions.
2Productivity
If the vehicle operates under degraded perception conditions with limited lane information, then the vehicle can continue moving autonomously, but the control accuracy and safety are reduced
Solution Approach 1:
The system applies partial action by using only the portion of the trajectory that can be reliably perceived. When perception is degraded, it plans and controls based on the available perception range rather than attempting full long-range trajectory control, thereby maintaining autonomous operation within safe limits.
Solution Approach 2:
The system maintains continuous autonomous operation by adapting to degraded perception conditions rather than stopping. It continuously adjusts the look-ahead point and trajectory planning to match available perception, ensuring uninterrupted but modified autonomous control.
3Reliability
If the system requests human driver intervention under degraded perception conditions, then safety is improved, but the autonomous operation is interrupted and productivity decreases
Solution Approach 1:
The system serves itself by autonomously adapting to degraded perception conditions through dynamic look-ahead point adjustment and modified trajectory planning. This self-service capability allows the system to handle degraded conditions without human intervention, maintaining both safety and autonomous operation continuity.
4Reliability
If the look-ahead point is moved closer to the vehicle under degraded perception, then the look-ahead point remains within perception range, but the trajectory planning accuracy and control precision are reduced
Solution Approach 1:
The system dynamically positions the look-ahead point based on perception range. When perception is degraded, the look-ahead point moves closer to maintain reliability; when perception improves, it moves farther to enhance accuracy. This dynamic positioning resolves the contradiction between perception compliance and planning accuracy.
Solution Approach 2:
The system uses feedback from perception range monitoring to adjust look-ahead point positioning. By continuously measuring perception quality and adjusting the look-ahead distance accordingly, the system maintains the optimal balance between staying within perception range and preserving trajectory planning accuracy.
Data Source
AI summary
Methods and systems of controlling a vehicle. The methods and systems include switching to a second processor executed mode in response to determining that a range of perception information extends to a range end point that is closer to the vehicle than a first Look Ahead (LA) point. In the second processor executed mode: a motion planner requests a lateral controller to provide a closer LA point relative to the vehicle. The lateral controller determines the closer LA point and sends the closer LA point to the motion planner. The lateral controller generates control commands for an EPS system based on an error between a desired trajectory and an actual trajectory at the closer LA point. The motion planner generates the desired trajectory based on perception data and the closer LA point.