Autonomous Vehicle Travel Corridor Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in navigating roadways shared with other vehicles and obstacles without clear lane markers, as existing systems struggle to effectively plan a safe and efficient corridor for travel.
Innovation Solution
A method and system for generating a corridor of travel that identifies the vehicle's current position, uses sensor data to adjust initial corridor seeding based on detected objects and road characteristics, and optimizes boundaries to minimize curvature differences while maintaining buffers, ensuring safe distance and avoiding collisions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the vehicle maintains a large buffer distance from objects on the roadway, then collision safety is improved, but the available travel corridor width is reduced
Solution Approach 1:
The buffer distance is made dynamic rather than static. The system adjusts the buffer distance based on relative velocity, object type, and environmental conditions. When objects are stationary or moving slowly, a smaller buffer is sufficient, allowing wider corridor usage. When objects are moving quickly or pose higher risk, the buffer automatically increases to maintain safety
Solution Approach 2:
The system changes the buffer parameter adaptively based on multiple factors including object classification (parked vs. moving vehicles), relative speed, curvature of the roadway, and sensor detection confidence. This allows the buffer distance to be optimized for each specific situation rather than using a fixed conservative value throughout
2Ease of operation
If the corridor curvature closely follows the roadway curvature, then navigation smoothness is improved, but the ability to avoid objects may be reduced
Solution Approach 1:
The corridor planning is divided into multiple segments along the roadway. Each segment's curvature is independently optimized to balance smoothness and safety. The system segments the roadway into zones based on object proximity and adjusts corridor curvature locally in each zone, allowing tight curves near objects and smoother curves in open areas
Solution Approach 2:
The corridor curvature is made dynamic and adaptive rather than following a fixed roadway curvature. The system continuously adjusts the corridor curvature based on real-time object detection, allowing the path to become sharper or gentler as needed to maintain both smooth navigation and effective obstacle avoidance
3Manufacturing precision
If the system generates multiple boundary options and selects the optimal one, then corridor optimization is improved, but computational complexity increases
Solution Approach 1:
The boundary generation process is segmented into discrete candidate boundaries at key locations along the roadway. Rather than generating continuous infinite possibilities, the system identifies a finite set of candidate boundaries based on object positions and roadway features, then selects the optimal combination, reducing computational complexity while maintaining optimization quality
Solution Approach 2:
The system performs preliminary actions by pre-identifying candidate boundary positions based on object detection and roadway geometry before final corridor selection. This preliminary structuring of possible boundaries reduces the search space and computational burden of finding the optimal corridor while still achieving high-quality results
Data Source
AI summary
A method includes identifying a current position of the vehicle, using sensor data; generating, via a processor, an initial corridor seeding based upon the current position of the vehicle and physical characteristics of the road, without consideration of any other objects on the roadway; identifying a plurality of objects along a first side of the roadway, the first side including a side of the roadway in which the vehicle is traveling, using the sensor data; generating, via the processor, a target first boundary for the corridor based on the identified plurality of objects along the first side of the roadway; generating, via the processor, a target second boundary on a second side of the roadway, opposite the first side; and adjusting, via the processor, the initial corridor seeding based on the target first and second boundaries, thereby generating the corridor of travel for the vehicle along the roadway.