Autonomous Vehicle Routing Using Local Routes and Remote Cost Planning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous vehicle routing systems require significant computing resources and time to generate full routes, which can hinder the vehicle's ability to react to changing roadway conditions, and implementing a centralized route planner may not efficiently utilize transient data.
Innovation Solution
A decentralized approach involving a local route planner and a general route planner, where the local planner generates routes from the vehicle's current location to local endpoints, and the general planner determines costs for these endpoints to a trip endpoint, allowing for faster reaction to changing conditions and reducing computational load on the vehicle.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a centralized route planner is used to generate full routes, then routing coverage is comprehensive, but computational resources and time consumption increase significantly
Solution Approach 1:
The routing system is divided into two independent planners: a local route planner that generates routes from current location to local endpoints, and a general route planner that determines costs from endpoints to trip destination. This segmentation allows each planner to operate independently with appropriate data, reducing overall computational time while maintaining comprehensive routing coverage.
Solution Approach 2:
The local route planner pre-generates routes to multiple potential endpoints before the vehicle reaches them. This preliminary action allows the system to have route options ready in advance, reducing the time needed for real-time route generation when conditions change.
2Loss of information
If a centralized route planner is used to generate full routes, then complete routing information is obtained, but computational load on the vehicle increases
Solution Approach 1:
The computational task is segmented between vehicle-based local planning and server-based general planning. The vehicle only needs to process local route generation and endpoint selection, while the server handles computationally intensive general route cost calculations, reducing the vehicle's computational load while maintaining information completeness.
Solution Approach 2:
A server acts as an intermediary between the vehicle and the general route planning system. The server receives endpoint information from the vehicle, calculates general route costs, and returns results to the vehicle, allowing the vehicle to obtain complete routing information without bearing the full computational burden.
3Speed
If the vehicle processes all routing computations locally, then routing decisions are made quickly, but computational resources are insufficient for complex calculations
Solution Approach 1:
Computational tasks are segmented by scope and timing: the local route planner handles time-critical local route generation using the vehicle's limited computational resources, while the general route planner handles less time-critical cost calculations using the server's abundant computational power.
Solution Approach 2:
The system performs preliminary local route generation at the vehicle level, which is computationally efficient and fast. This preliminary action provides immediate route options without requiring intensive computations, maintaining speed while managing computational power constraints.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Various examples are directed to systems and methods for controlling an autonomous vehicle. For example, a navigator system at an autonomous vehicle may generate a plurality of local routes beginning at a vehicle location and extending to a plurality of local route end points. The navigator system may access general route cost data, the general route cost data describing general route costs from the plurality of local route end points to a trip end point. The navigator system may select the first local route of the plurality of routes based at least in part on the general route cost data. A vehicle autonomy system at the autonomous vehicle may begin to control the autonomous vehicle along the first local route.