Constrained Smoothing Spline Speed Optimization for Autonomous Vehicles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous driving vehicles face challenges in navigating smoothly and efficiently due to the lack of effective speed optimization methods that consider obstacles and traffic information, leading to unstable vehicle control and potential collisions.
Innovation Solution
The implementation of a constrained smoothing spline approach using quadratic programming optimization to generate smooth reference lines and speed trajectories, which takes into account boundary areas, road boundaries, and obstacles, ensuring stable and collision-free navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional path planning methods are used without speed optimization, then the vehicle can navigate obstacles, but the vehicle control becomes unstable and collisions may occur
Solution Approach 1:
The patent optimizes speed parameters along the reference line using smoothing spline interpolation. By continuously adjusting the speed profile based on curvature, distance to obstacles, and traffic information, the system achieves stable vehicle control while maintaining efficient navigation. The speed optimization transforms discrete speed points into a smooth continuous speed trajectory.
2Reliability
If smooth reference lines are generated using conventional methods, then vehicle control stability improves, but the computation time increases and real-time performance deteriorates
Solution Approach 1:
The patent pre-generates smooth reference lines using quadratic programming optimization before real-time navigation. The reference line is computed in advance considering road boundaries and static obstacles, reducing real-time computation requirements. During navigation, the system only needs to optimize speed along the pre-computed path, significantly reducing computation time while maintaining stability.
3Reliability
If speed optimization considers all obstacles and traffic information, then collision risk reduces, but the device complexity increases
Solution Approach 1:
The patent segments the navigation problem into two independent parts: path planning (reference line generation) and speed optimization. The reference line considers static obstacles and road boundaries, while speed optimization separately considers dynamic obstacles and traffic information. This segmentation reduces system complexity by avoiding the need to simultaneously process all constraints in a single complex optimization problem.
4Productivity
If aggressive speed profiles are used for efficient navigation, then productivity increases, but vehicle control stability deteriorates and collision risk increases
Solution Approach 1:
The patent implements dynamic speed optimization that continuously adapts the speed profile based on real-time conditions. The smoothing spline adjusts speed dynamically along the reference line, reducing speed in high-curvature sections and near obstacles while maintaining higher speeds in safe, straight sections. This dynamic adjustment achieves both navigation efficiency and control stability.
Data Source
AI summary
According to some embodiments, a system selects a number of polynomials representing a number of time segments of a time duration to complete the path trajectory. The system selects an objective function based on a number of cost functions to smooth speeds between the time segments. The system defines a set of constraints to the polynomials to at least ensure the polynomials are smoothly joined together. The system performs a quadratic programming (QP) optimization on the objective function in view of the set of constraints, such that a cost associated with the objective function reaches a minimum while the set of constraints are satisfied. The system generates a smooth speed for the time duration based on the optimized objective function to control the ADV autonomously.