Vehicle Trajectory Planning Using Driving-Mode-Aware Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing intelligent vehicles struggle with accurately predicting the trajectories of surrounding vehicles due to the uncertainty in vehicle motion, especially when autonomous and manual driving vehicles coexist, leading to challenges in deterministic trajectory planning and increased collision risks.
Innovation Solution
An intelligent driving domain controller that utilizes communication technology to obtain and analyze the driving mode and trajectory of surrounding vehicles, employing customized trajectory prediction models based on historical driving data and real-time parameters to ensure accurate trajectory planning, including manual and autonomous modes, and ensures secure data transmission.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If trajectory prediction is performed based only on current moment driving condition using unified rules, then the processing complexity is reduced, but the prediction accuracy deteriorates due to uncertainty in vehicle motion and diverse driving behaviors
Solution Approach 1:
The system dynamically adapts the trajectory prediction approach based on the identified driving mode. For autonomous vehicles, it uses complex motion model prediction that accounts for vehicle dynamics and control characteristics. For manual vehicles, it employs driver behavior pattern recognition. This dynamic adaptation allows the system to use appropriate complexity levels for different scenarios, improving prediction accuracy without unnecessarily increasing processing complexity for all cases.
Solution Approach 2:
The system changes the prediction parameters and models based on the identified driving mode. When autonomous driving mode is detected, it uses parameters related to automated control systems and motion planning. When manual driving mode is detected, it switches to parameters related to driver behavior patterns and subjective decision-making. This parameter adaptation enables accurate prediction across diverse driving scenarios while managing computational complexity.
2Measurement precision
If customized trajectory prediction models based on historical data and driving habits are employed, then the prediction accuracy is improved, but the device complexity and data processing requirements increase
Solution Approach 1:
The system segments the trajectory prediction task into distinct components based on driving mode: autonomous vehicle prediction and manual vehicle prediction. Each segment uses specialized models tailored to its characteristics. This segmentation allows the system to manage complexity by handling different prediction types separately rather than using a single complex model for all cases.
Solution Approach 2:
The system performs preliminary classification of driving modes before executing the actual trajectory prediction. By first identifying whether a vehicle is autonomous or manual through communication and pattern recognition, the system can prepare and apply the appropriate prediction model in advance. This preliminary action simplifies the overall process by avoiding the need to run multiple complex models simultaneously.
3Measurement precision
If real-time communication technology is used to obtain surrounding vehicle trajectories, then the trajectory planning accuracy is improved, but the data transmission security requirements and system complexity increase
Solution Approach 1:
The system uses communication technology as an intermediary to obtain trajectory information from surrounding vehicles. Rather than directly accessing or interfering with vehicle systems, it receives trajectory data through standardized communication interfaces. This intermediary approach enables accurate trajectory planning while maintaining security boundaries and reducing direct system complexity.
Data Source
AI summary
This application provides a method and an apparatus for planning a vehicle trajectory, an intelligent driving domain controller, and an intelligent vehicle. One example method includes: An intelligent driving domain controller of a first vehicle obtains a first trajectory of the first vehicle, obtains a second trajectory of at least one second vehicle based on a first communications technology, and then determines trajectory planning of the first vehicle based on the first trajectory and the second trajectory of the at least one second vehicle.