Autonomous Vehicle Trajectory Re-Linearization on Evolving Reference Paths
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional autonomous vehicle path planning systems in high-dimensional Cartesian coordinate frames fail to efficiently decouple lateral and longitudinal components, making it difficult to optimize path planning and navigate complex environments.
Innovation Solution
Decompose path planning into lower-dimensional curvilinear coordinate frames to linearize vehicle dynamics, allowing for the decoupling of lateral and longitudinal components, and utilize convex optimization techniques to refine trajectories iteratively.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If path planning is performed in high-dimensional Cartesian coordinate frames, then the system can represent complex trajectories, but it fails to efficiently decouple lateral and longitudinal components
Solution Approach 1:
The patent segments the path planning problem by transforming from a unified high-dimensional Cartesian coordinate system into separate curvilinear coordinate frames that independently represent lateral and longitudinal dimensions. This segmentation allows each component to be optimized separately, resolving the contradiction between ease of operation and device complexity.
Solution Approach 2:
The patent applies dimensionality change by transitioning from high-dimensional Cartesian coordinates to lower-dimensional curvilinear coordinates. This transformation reduces the effective dimensionality of the planning problem while maintaining the ability to represent complex trajectories, thereby simplifying the decoupling of motion components.
2Productivity
If the trajectory is linearized about a static reference path, then computational speed improves, but it fails to account for evolving optimal paths during execution
Solution Approach 1:
The patent makes the reference path dynamic by continuously updating it during trajectory execution based on changing conditions and optimization results. This dynamic update allows the system to maintain computational efficiency through linearization while simultaneously adapting to evolving optimal paths, resolving the contradiction between productivity and adaptability.
Solution Approach 2:
The patent implements feedback by using the executed trajectory and environmental information to update the reference path for subsequent linearization steps. This feedback mechanism ensures that the system adapts to evolving conditions while maintaining computational efficiency through iterative refinement.
3Manufacturing precision
If convex optimization is used to refine trajectories, then optimization accuracy improves, but computational complexity increases
Solution Approach 1:
The patent changes the parameterization of the trajectory problem by using curvilinear coordinates and breaking the trajectory into segments. This parameter transformation converts the original complex optimization problem into a series of simpler convex optimization problems that can be solved efficiently while maintaining high accuracy.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A system of linearizing a trajectory of an autonomous vehicle about a reference path includes a computing device and a computer-readable storage medium. The computer-readable storage medium includes one or more programming instructions that, when executed, cause the computing device to receive a reference path for an autonomous vehicle, where the reference path defines a proposed trajectory for the autonomous vehicle in a Cartesian reference frame, identify an objective based on the received reference path, where the objective comprises a longitudinal component and a lateral component, project the objective into a curvilinear coordinate frame described by the received reference path, decouple the longitudinal component and the lateral component, linearize the lateral component about the reference path, generate a new reference path for the autonomous vehicle by fusing the linearized longitudinal component and the linearized lateral component, and map the new reference path back to the Cartesian reference frame.