Driving Difficulty Heat Maps for Autonomous Route Disengagement Risk
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in navigating routes that require consideration of disengagement likelihoods due to factors like lane changes, unprotected turns, and traffic conditions, which existing routing systems do not adequately address.
Innovation Solution
A method and system for generating a driving difficulty heat map using a disengage model trained with log data from manual driving mode, which predicts the likelihood of disengagement from autonomous driving mode by assigning outputs to a grid of cells, allowing for the identification of potentially difficult areas and routes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If routing systems use traditional shortest distance or shortest time metrics, then route efficiency is improved, but the likelihood of vehicle disengagement increases due to difficult driving conditions
Solution Approach 1:
The patent changes the routing parameters by incorporating disengagement probability metrics into the route selection criteria. Instead of only considering distance or time, the system evaluates routes based on predicted disengagement likelihood at each segment, allowing it to balance efficiency with safety by selecting routes that minimize overall disengagement risk while maintaining reasonable travel time.
Solution Approach 2:
The patent introduces an intermediary component - the disengage model - that acts as a mediator between the routing system and the vehicle control system. This model predicts disengagement probability for each route segment based on environmental factors, traffic conditions, and maneuver complexity, providing an additional layer of assessment that informs route planning decisions without directly controlling the vehicle.
2Adaptability or versatility
If autonomous vehicles operate in complex driving environments requiring frequent disengages, then adaptability to various conditions is improved, but operational reliability deteriorates
Solution Approach 1:
The patent applies preliminary action by training the disengage model in advance using extensive log data from manual and autonomous driving. This pre-trained model enables the system to predict disengagement scenarios before they occur during actual autonomous operation, allowing the routing system to proactively avoid high-risk segments rather than reacting to problems as they arise.
Solution Approach 2:
The patent implements feedback mechanisms where the disengage model continuously evaluates route segments and provides probability assessments that feed back into the routing system. This closed-loop feedback allows the system to learn from actual disengagement events and refine its predictions, improving both adaptability to new conditions and overall reliability through iterative optimization.
Data Source
AI summary
Aspects of the disclosure provide for the generation of a driving difficulty heat map for autonomous vehicles. For instance, log data generated by a vehicle being driven in a manual driving mode for a segment of a route may be input into a disengage model in order to generate an output identifying a likelihood of a vehicle driving in an autonomous driving mode requiring a disengage from the autonomous driving mode along the segment of the route. The log data may have been collected within a geographic area. A grid for the geographic area may be generated. The grid may include a plurality of cells. The output is assigned to one of the plurality of cells. The plurality of cells and assigned output may be used to generate a driving difficulty heat map for the geographic area.