Autonomous Vehicle Route Scoring Using Predicted Scene Difficulty
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current navigation systems for autonomous vehicles primarily consider distance and driving time, offering limited options and failing to account for the difficulty of road segments, which can lead to unpredictable driving experiences due to factors like obstacles, occlusions, and time-dependent challenges.
Innovation Solution
A method and system that determine a route for autonomous vehicles by computing a route cost based on difficulty values, which include factors such as disengagement from autonomous mode, number of unprotected turns, occlusions, and obstacles, using data from vehicle logs to weight time, distance, and difficulty scores, allowing for dynamic adjustments based on passenger input or scenarios.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If navigation systems only consider distance and driving time, then route selection is simple and fast, but the driving experience becomes unpredictable and unsafe due to difficulty factors
Solution Approach 1:
The system pre-computes difficulty values for road segments using historical data from vehicle logs before actual route selection. This preliminary action stores difficulty information (including disengagements, requests for assistance, unprotected turns, occlusions, and obstacles) so that when route selection occurs, the system can quickly evaluate multiple routes using pre-calculated difficulty metrics rather than computing everything in real-time, thus improving safety without excessive complexity
Solution Approach 2:
The patent introduces difficulty values as an intermediary metric that bridges the gap between simple navigation parameters (distance, time) and complex safety considerations. These difficulty values aggregate multiple safety factors into a single comparable metric, allowing the route selection system to consider safety without directly managing the complexity of individual difficulty factors
2Reliability
If the system computes route cost based on multiple difficulty factors, then route safety improves, but computational complexity increases
Solution Approach 1:
Difficulty values are pre-computed and stored in a database using historical vehicle log data before actual route selection occurs. This preliminary computation separates the heavy processing burden from real-time operation, allowing the system to maintain high route safety through comprehensive difficulty factor analysis while keeping real-time computational power requirements manageable
Solution Approach 2:
The system uses weighted scores where not all difficulty factors are equally weighted in every situation. By adjusting weights based on passenger input or specific scenarios, the system can focus computational resources on the most relevant difficulty factors for each particular route selection, reducing overall computational power requirements while maintaining safety
3Adaptability or versatility
If the system uses fixed route selection criteria, then processing is fast, but adaptability to different passenger preferences and scenarios is limited
Solution Approach 1:
The system dynamically adjusts the weights of different difficulty factors based on passenger input or specific scenarios. Instead of fixed criteria, the weighting scheme can change to prioritize different factors (e.g., safety vs. time) depending on passenger preferences, making the system adaptable while maintaining ease of operation through automated weight adjustment rather than requiring complex manual configuration
Solution Approach 2:
The system incorporates feedback mechanisms where passenger input or scenario information influences the weighting of difficulty factors. This feedback loop allows the system to adapt to different preferences and scenarios while keeping the interface simple, as the complexity of adaptation is handled automatically through the feedback-driven weight adjustment process
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A route is selected for travel by an autonomous vehicle based on at least a level of difficulty of traversing the driving environment along that route. Vehicle signals, provided by one or more autonomous vehicles, indicating a difficulty associated with traveling a portion of a route are collected and used to predict a most favorable driving route for a given time. The signals may indicate a probability of disengaging from autonomous driving mode, a probability of being stuck for an unduly long time, traffic density, etc. A difficulty score may be computed for each road segment of a route, and then the scores of all of the road segments of the route are added together. The scores are based on number of previous disengagements, previous requests for remote assistance, unprotected left or right turns, whether parts of the driving area are occluded, etc. The difficulty score is used to compute a cost for a particular route, which may be compared to costs computed for other possible routes. Based on such information, a route may be selected.