Autonomous Vehicle Dispatch Using Learned Routing Constraints
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous vehicle dispatch systems face challenges in accurately determining routing constraints for autonomous vehicles due to incomplete or incorrect vehicle capability data, leading to inefficiencies and potential safety issues when selecting vehicles for trips.
Innovation Solution
A dispatch system that learns from trip result data to determine and apply routing constraints, adjusting the connectivity and cost of route components in the routing graph to better reflect the capabilities and performance of different autonomous vehicle types.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If vehicle capability data is used to determine routing constraints, then vehicle selection can be automated, but the accuracy and reliability of routing constraints deteriorate due to incomplete or incorrect data
Solution Approach 1:
The system implements feedback by collecting actual trip result data from autonomous vehicle operations and using this real-world performance information to iteratively refine and update routing constraints. This closed-loop approach allows the system to learn from actual vehicle performance and continuously improve the accuracy of routing decisions, resolving the contradiction between automation and reliability.
Solution Approach 2:
The system performs preliminary action by proactively collecting and analyzing trip result data to identify patterns in vehicle performance before making routing decisions. By pre-processing this performance data to establish empirical routing constraints, the system ensures that automated vehicle selection is based on validated performance patterns rather than incomplete manufacturer specifications.
2Measurement precision
If routing constraints are adjusted based on real-world performance data, then vehicle-route matching accuracy improves, but system complexity increases
Solution Approach 1:
The system applies universality by using a multi-functional data processing module that handles multiple tasks: collecting trip result data, analyzing vehicle performance patterns, updating routing constraints, and validating vehicle capabilities. This consolidated approach improves measurement precision while avoiding the need for separate complex systems for each function, thereby managing overall system complexity.
3Manufacturing precision
If trip result data is collected and analyzed to refine routing constraints, then vehicle selection accuracy improves, but data processing time and computational resources increase
Solution Approach 1:
The system performs preliminary action by continuously pre-processing and storing trip result data as it becomes available, organizing it into structured formats that facilitate rapid analysis. This proactive data preparation ensures that when routing decisions need to be made, the system can quickly query and apply relevant performance patterns without experiencing delays from real-time data processing, thus improving vehicle selection accuracy while minimizing time loss.
Data Source
AI summary
Various examples described herein are directed to systems and methods for dispatching trips to a plurality of autonomous vehicles. For example, a dispatch system may access trip result data describing a plurality of trips executed by a set of autonomous vehicles of a first vehicle type. The dispatch system may determine a first routing constraint for autonomous vehicles of the first vehicle type using the trip result data and select an autonomous vehicle to execute a first new trip using the first routing constraint. The dispatch system may send a request to execute the first new trip to the selected autonomous vehicle.