Motion Primitive Trajectory Planning for Autonomous Vehicles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for guiding autonomous vehicles through environments are computationally intensive, especially when dealing with dynamic curvatures, making them inefficient in resource-limited or time-limited scenarios.
Innovation Solution
The use of approximate motion primitives, which are determined by parsing reference trajectories into segments with positive, negative, or zero curvature, and parameterizing acceleration as piece-wise constant and lateral offset as a cubic function, allowing for faster and more efficient trajectory planning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If computationally intensive coordinate transformations or boundary value problems are used to generate routes, then route planning accuracy is improved, but processing time and computational resource consumption increase
Solution Approach 1:
The reference trajectory is divided into multiple segments with different curvature characteristics (positive curvature, negative curvature, zero curvature). This segmentation allows the system to apply simplified motion primitive models to each segment rather than using computationally intensive methods for the entire trajectory, thereby reducing processing time while maintaining planning accuracy.
Solution Approach 2:
The patent changes the parameterization approach by representing trajectories using motion primitives with specific curvature parameters (κ₀, κ₁) and polynomial coefficients (a₀, a₁, a₂, a₃) instead of traditional coordinate transformations. This parameter change enables faster computation while preserving the essential geometric characteristics needed for accurate route planning.
2Measurement precision
If computationally intensive coordinate transformations or boundary value problems are used to generate routes, then route planning accuracy is improved, but computational resource consumption increases
Solution Approach 1:
By segmenting the trajectory into curvature-based sections, the system avoids applying complex computational algorithms uniformly across the entire path. Each segment can be processed independently using simpler motion primitive calculations, reducing overall computational resource consumption while maintaining planning accuracy.
Solution Approach 2:
The patent uses approximate motion primitive models that are computationally inexpensive compared to traditional methods. These simplified models provide sufficient accuracy for route planning while consuming significantly fewer computational resources, effectively replacing expensive calculations with cheaper alternatives.
3Productivity
If traditional trajectory planning methods are used, then comprehensive trajectory coverage is achieved, but processing speed decreases
Solution Approach 1:
The system dynamically adapts its processing approach based on the curvature characteristics of each trajectory segment. By identifying whether a segment has positive, negative, or zero curvature, the system can select appropriate motion primitive models and parameterization strategies, enabling fast processing while maintaining versatility in handling different trajectory types.
Solution Approach 2:
Different motion primitive models and parameterization methods are applied to different types of trajectory segments based on their local curvature characteristics. This local quality approach ensures that each segment is processed with the most appropriate method, achieving both high processing speed and comprehensive trajectory coverage.
Data Source
AI summary
In autonomous driving, it is often useful to plan trajectories in a curvilinear coordinate frame with respect to some reference trajectory, like a path produced by a hi-level route planner. This disclosure includes techniques for developing efficient approximate path coordinate motion primitives appropriate for fast planning in autonomous driving scenarios. These primitives are approximate in that particular quantities, like the path length, acceleration, and track offset trajectory, are known with some degree of certainty, and values that depend on the curvature of the reference path can be bound. Such approximate motion primitives can be used to control the autonomous vehicle to follow the trajectory in an environment.