Autonomous Vehicle Route Prediction Using Difference Data

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Assigning transportation services to autonomous vehicles (AVs) is challenging due to differences in capabilities and routing preferences among various AVs, making it difficult for the service assignment system to select the best AV for a service, especially when routing engines separate from the system are involved.

Innovation Solution

The service assignment system uses route prediction and difference data to determine acceptable routes, modifying routing rules to favor or disfavor certain roadway elements based on common properties, statuses, and risks, thereby improving the selection of AVs for transportation services.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If the service assignment system uses a routing engine separate from the service assignment system to route AVs, then the AVs can have more flexible routing capabilities and preferences, but it becomes more difficult for the service assignment system to select the best AVs for transportation services

Engineering Contradiction:

Improverouting capabilitiesVSAvoidservice assignment complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system performs preliminary routing using the service assignment system's routing engine to generate predicted routes for candidate AVs before final service assignment. This preliminary routing action allows the system to evaluate AVs based on predicted route quality while still allowing AVs to provide their own planned routes, thus maintaining flexibility without losing selection control.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system compares the predicted routes generated by the service assignment system with the actual planned routes provided by the AVs. This feedback mechanism allows the system to assess the quality of AV routing decisions and use this information to improve future service assignments, resolving the complexity issue while preserving routing flexibility.

Inventive Principle:

Principle #23Feedback

2Measurement precision

If the service assignment system compares predicted routes with planned routes to determine acceptability, then the selection accuracy of AVs improves, but the computational time and processing complexity increases

Engineering Contradiction:

ImproveAV selection accuracyVSAvoidservice assignment time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system performs partial comparison by focusing on key route characteristics and differences between predicted and planned routes rather than analyzing every detail. The route difference engine identifies significant deviations and evaluates them against routing rules, providing sufficient accuracy for service assignment without requiring complete route analysis, thus reducing processing time.

Inventive Principle:

Principle #16Partial or excessive action

3Reliability

If the system modifies routing rules based on difference data between predicted and planned routes, then the routing quality improves over time, but the system complexity and difficulty of managing routing rules increases

Engineering Contradiction:

Improverouting qualityVSAvoidrouting rule management

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system automatically modifies routing rules based on difference data collected from comparing predicted and planned routes. The route difference engine identifies patterns in route deviations and autonomously adjusts routing rules without requiring manual intervention, thus improving routing quality while keeping the system manageable through self-optimization.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system replaces manual routing rule management with an automated machine learning approach. The route difference engine uses algorithms to analyze difference data and automatically adjust routing parameters, substituting mechanical rule management with intelligent automated systems that improve routing quality without increasing operational complexity.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS20220065647A1Autonomous vehicle planned route prediction

Publication Date: 2022.03.03 UBER TECHNOLOGIES INC

AI summary

Various examples are directed to systems and methods for providing transportation services. A service assignment system may receive a transportation service request from a user. The transportation service request may describe a transportation service having a start location and an end location. The service assignment system may select a first autonomous vehicle (AV) of a first AV type and determine a first predicted route for the first AV using vehicle capability data describing the first AV type and first difference data describing a difference between a previous predicted route for a previous AV of the first AV type and a previous planned route received from the previous AV of the first AV type. The service assignment system may receive, from the first AV, a first planned route for executing the transportation service and instruct the first AV to begin executing the transportation service.