Autonomous Vehicle Trajectory Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in smoothly transitioning between baseline and changed trajectories, leading to potential jerks or collisions due to inefficient path adjustments.
Innovation Solution
A method involving a computer system that defines vertices on the baseline trajectory, assigns cost functions based on distance and other criteria, and optimizes these vertices to determine a transitional trajectory that smoothly transitions from the baseline to the new trajectory, ensuring safe and efficient navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If the vehicle directly switches from baseline trajectory to changed trajectory, then the response time is reduced, but the vehicle stability deteriorates causing jerks or collisions
Solution Approach 1:
The system performs preliminary action by pre-calculating and optimizing a transitional trajectory before the vehicle actually transitions. The computer system defines vertices on the baseline trajectory and pre-computes optimal transition paths to the changed trajectory, ensuring that when transition is needed, a smooth pre-planned path is already available. This eliminates abrupt changes while maintaining rapid response.
Solution Approach 2:
The system applies dynamics by making the trajectory adaptable and adjustable during operation. The transitional trajectory is dynamically optimized based on real-time conditions, with vertices and path parameters that can be adjusted to balance speed and smoothness. This allows the vehicle to transition efficiently while maintaining stability through continuous adaptation of the path parameters.
2Stability of the object's composition
If the vehicle follows a smooth transitional trajectory, then the vehicle stability is maintained, but the time to reach the new trajectory increases
Solution Approach 1:
The system utilizes parameter changes by optimizing vertices and trajectory parameters to achieve the best balance between smoothness and speed. The computer system adjusts parameters such as vertex positions, transition timing, and path curvature to minimize transition time while maintaining vehicle stability. This allows the transitional trajectory to be neither too abrupt nor too conservative.
3Device complexity
If simple trajectory adjustment is used, then the computational complexity is reduced, but the trajectory optimization quality deteriorates
Solution Approach 1:
The system applies segmentation by dividing the trajectory into discrete vertices along the baseline trajectory. Each vertex can be independently optimized with cost functions, allowing complex optimization to be broken down into manageable segments. This reduces computational complexity compared to optimizing the entire trajectory as a single continuous path, while still achieving high optimization quality through localized vertex adjustments.
Solution Approach 2:
The system uses parameter changes by defining cost functions that evaluate and optimize specific trajectory parameters such as distance to the changed trajectory, smoothness metrics, and safety constraints. These parameter-based cost functions enable systematic optimization of trajectory quality without requiring overly complex computational models, achieving good balance between computational efficiency and optimization quality.
Data Source
AI summary
A vehicle configured to operate in an autonomous mode is provided. The vehicle is configured to follow a baseline trajectory. Changes to the baseline trajectory are received by a computer system associated with the vehicle. The computer system creates a function representing the current trajectory of the vehicle, as well as one or more functions defining any desired trajectory criteria, and generates an optimization problem from the functions, which, when solved, provide a new trajectory for the vehicle to follow that moves efficiently and smoothly toward the changed baseline trajectory.