Vehicle Path Planning With Previous-Path Constraints
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current path planning systems for autonomous and semi-autonomous vehicles often result in erratic vehicle motion due to significant lateral displacement of the target path from the previous path, leading to uncomfortable maneuvers for occupants.
Innovation Solution
A method for path planning that incorporates constraints based on the previous path, using quadratic programming to generate a new path within defined lateral boundaries, which are dynamically adjusted based on vehicle characteristics and environmental factors, ensuring smoother and more stable navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional path planning generates a target path from a given drivable area, then the vehicle can navigate through the environment, but the updated path may be laterally displaced from the previous path causing erratic vehicle motion
Solution Approach 1:
The patent applies preliminary action by pre-defining lateral boundaries based on the previous path before generating the updated target path. These boundaries are established in advance to constrain the optimization process, ensuring that the new path remains within acceptable lateral displacement limits from the previous path, thereby preventing erratic vehicle motion while maintaining adaptability.
2Measurement precision
If the target path is updated to improve navigation accuracy, then the vehicle can follow a more precise route, but the lateral displacement from the previous path causes uncomfortable maneuvers for occupants
Solution Approach 1:
The patent applies parameter changes by dynamically adjusting the lateral boundary parameters based on the previous path characteristics. The optimization process modifies the path parameters within these dynamically adjusted boundaries, achieving a balance between path accuracy and smooth transitions. This ensures that the vehicle follows a precise route while maintaining comfortable maneuvers for occupants by limiting excessive lateral displacements.
3Speed
If the path is updated frequently to respond to environmental changes, then the vehicle can adapt to new conditions, but the frequent updates increase the risk of generating erratic paths
Solution Approach 1:
The patent applies feedback by using the previous path as a constraint reference for generating the updated path. The lateral boundaries are defined based on the historical path information, creating a feedback mechanism that ensures each new path update remains consistent with previous trajectories. This feedback loop maintains path consistency even during frequent updates in response to environmental changes, reducing the risk of erratic path generation.
Data Source
Figure 1
Figure 2
Figure 3(a)~3(c)
AI summary
The present disclosure relates to a method (400, 500) for path planning for an autonomous or semi-autonomous vehicle. The method comprises obtaining (402a-d) a drivable area of a surrounding environment of the vehicle, and generating (403a-d, 502) a path (3) within the drivable area for a time step t based on a predefined set of characteristics for the path and a predefined set of constraints. The predefined set of constraints comprise at least one constraint (7a, 7b, 8a, 8b) based on a path (2) generated for a previous time step.