Autonomous Vehicle Route Planning for Unsafe Road Segment Screening
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in safely navigating certain road segments and conditions due to limitations in sensor capabilities and software processes, which can lead to degraded performance or failure in perceiving the environment, performing motion planning, and making decisions, especially under varying weather conditions and complex scenarios.
Innovation Solution
The implementation of a route planning system that analyzes the performance of autonomous vehicles on road features, considering prior driving and simulation data, to determine which routes can be safely and robustly traveled by eliminating unsuitable road segments based on sensor and software capabilities, environmental conditions, and road feature characteristics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the autonomous vehicle attempts to navigate all possible road segments to maximize route options, then the coverage and versatility of the vehicle is improved, but the safety and reliability deteriorate due to sensor and software limitations in complex or adverse conditions
Solution Approach 1:
The system performs preliminary analysis of road segment safety characteristics before the vehicle attempts to navigate them. Road segments are pre-evaluated based on sensor capability requirements, environmental conditions, and historical performance data, and are categorized as safe or unsafe for autonomous navigation. This preliminary classification allows the routing algorithm to exclude unsafe segments in advance, ensuring safety while maintaining route coverage.
2Reliability
If the vehicle excludes road segments with uncertain safety to maintain high safety standards, then the reliability is improved, but the number of available routes and adaptability decreases
Solution Approach 1:
The system applies different safety thresholds and evaluation criteria to different road segments based on their local characteristics. Rather than uniformly excluding all uncertain road segments, the system evaluates each segment individually considering factors such as sensor requirements, environmental conditions, traffic complexity, and historical performance. This localized quality assessment allows the vehicle to navigate safe uncertain segments while avoiding unsafe ones, maintaining both safety and route adaptability.
3Reliability
If the routing algorithm considers detailed sensor and software performance constraints to ensure safe navigation, then the safety is improved, but the computational complexity and device complexity increases
Solution Approach 1:
The system pre-computes and stores safety characteristics of road segments in a database, including sensor capability requirements, environmental condition thresholds, and historical performance metrics. This preliminary computation transforms complex real-time safety evaluation into simpler queries during route planning, reducing computational complexity while maintaining comprehensive safety considerations.
Solution Approach 2:
The system introduces an intermediate layer of abstraction between the detailed sensor/software constraints and the routing algorithm. Safety characteristics are represented as standardized parameters and categories that the routing algorithm can process efficiently, rather than directly evaluating complex sensor and software performance constraints. This intermediary representation simplifies the routing decision-making process while ensuring safety requirements are met.
Data Source
AI summary
Among other things, a determination is made of and ability of an autonomous vehicle to safely or robustly travel a road feature or a road segment or a route that is being considered for the autonomous vehicle as of a time or range of times. The route conforms to properties of stored road network information. The road feature or road segment or route is eliminated from consideration if the computer has determined that the road feature or road segment or route cannot be safely or robustly traveled by the autonomous vehicle. The determination is based on analysis of performance of the autonomous vehicle.