Autonomous Vehicle Obstacle Trajectory Prediction Using Polynomial Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional autonomous driving systems use fixed exponential curves to predict obstacle trajectories, which are not aligned with the obstacle's heading direction and do not consider lane shape, leading to potential safety and comfort issues for passengers.
Innovation Solution
The system divides trajectory prediction into longitudinal and lateral movement trajectories, optimized using polynomial functions (quintic for lateral and quartic for longitudinal) to ensure smooth alignment with the obstacle's heading direction and lane shape, generating a more accurate final predicted trajectory.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional systems use fixed exponential curves to predict obstacle trajectories, then the prediction process is simple, but the trajectory is not aligned with the obstacle's heading direction and does not consider lane shape, leading to safety issues
Solution Approach 1:
The trajectory prediction is divided into two independent segments: longitudinal trajectory prediction and lateral trajectory prediction. The longitudinal trajectory considers the obstacle's heading direction and uses polynomial fitting to generate smooth paths. The lateral trajectory incorporates lane shape information and constraints to ensure the predicted path follows the lane geometry. This segmentation allows each component to be optimized independently while maintaining overall prediction accuracy and safety.
2Measurement precision
If polynomial functions are used to optimize trajectories, then the alignment with heading direction and lane shape is improved, but the computational complexity increases
Solution Approach 1:
The system changes the mathematical parameters used for trajectory representation from fixed exponential curves to polynomial functions with adjustable coefficients. By fitting polynomials to the desired trajectory points and optimizing the coefficients to satisfy constraints (heading direction alignment, lane shape following), the system achieves precise trajectory alignment. The polynomial order and constraint parameters are adjusted to balance precision requirements with computational efficiency.
Data Source
AI summary
According to one embodiment, an obstacle is predicted to move from a starting point to an end point based on perception data perceiving a driving environment surrounding an ADV that is driving within a lane. A longitudinal movement trajectory from the starting point to the end point is generated in view of a shape of the lane. A lateral movement trajectory from the starting point to the end point is generated, including optimizing a shape of the lateral movement trajectory using a first polynomial function. The longitudinal movement trajectory and the lateral movement trajectory are then combined to form a final predicted trajectory that predicts how the obstacle is to move. A path is generated to control the ADV to move in view of the predicted trajectory of the obstacle, for example, to avoid the collision with the obstacle.