Shared Vehicle Control Using Driver Trajectory Risk Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing shared control methods for autonomous driving vehicles fail to adequately consider the driver's own driving behavior, leading to safety risks when the vehicle's risk of forward collision is high and the automatic system relies solely on autonomous control.
Innovation Solution
A trajectory-prediction-based shared control method that predicts the driver's trajectory, evaluates the driving risk, and dynamically adjusts the control weight distribution between the human driver and the automatic system based on the assessed risk.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the system deprives the driver of control when forward collision risk is high, then autonomous control reliability is improved, but the driver's ability to intervene safely is lost
Solution Approach 1:
The control weight distribution between autonomous system and driver is dynamically adjusted based on real-time risk assessment. When risk is high, the system automatically increases autonomous control weight; when risk is low and driver behavior is safe, driver control weight is increased. This dynamic adjustment resolves the contradiction by making the system reliable when needed while preserving driver adaptability when safe.
Solution Approach 2:
The system changes the parameter of control weight distribution based on risk level and driver behavior assessment. By monitoring driver trajectory predictions and risk metrics, the system adjusts the control weight parameter to balance autonomous reliability with driver intervention capability, preventing complete deprivation of driver control unless absolutely necessary.
2Measurement precision
If the system relies solely on autonomous control when risk is high, then control precision is improved, but the driver's safety intervention ability deteriorates
Solution Approach 1:
The system continuously monitors driver behavior through trajectory prediction and risk assessment, providing feedback on driver performance. This feedback mechanism allows the system to maintain high control precision through autonomous control while simultaneously evaluating whether the driver should be permitted to intervene, thus preserving safety intervention ability when appropriate.
Solution Approach 2:
The system performs preliminary risk assessment and driver behavior evaluation before completely depriving driver control. By predicting driver trajectory and assessing risk in advance, the system ensures that autonomous control precision is maintained while preventing situations where the driver would be completely unable to intervene if safe conditions arose.
3Device complexity
If the control weight is fixed based on current vehicle state, then system complexity is reduced, but the ability to predict and respond to future risks deteriorates
Solution Approach 1:
The system performs preliminary trajectory prediction and risk assessment based on driver behavior patterns before finalizing control weight distribution. This preliminary action allows the system to anticipate future risks and adjust control weights proactively, enhancing prediction capability without significantly increasing system complexity through efficient algorithms.
Solution Approach 2:
The system uses driver's own historical driving behavior data to predict future trajectory and assess risk, allowing the system to self-evaluate driver performance without requiring complex external assessment mechanisms. This self-service approach maintains prediction capability while controlling system complexity.
Data Source
AI summary
The present invention relates to the field of intelligent vehicles, and more specifically, to a human-machine shared control method based on trajectory prediction. It comprises: establishing a road-vehicle model based on vehicle-related parameters and then designing a path-tracking controller; further predicting the driver's driving trajectory, evaluating the driving risk, and switching the control power between the automatic system and the human driver based on the driving risk. The present invention ensures that when the driver's inputs will cause high risk to the vehicle, the driver's input is deprived of; and that when the driver's driving inputs will not cause high risk to the vehicle, the driving is carried out by the driver, so that in the case where the automatic system is having a problem, the driver can effectively realize a safe intervention, and the safety of the intelligent driving is improved.