Autonomous Vehicle Roundabout Lane Selection Using Occupancy Cost Functions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous vehicle navigation systems fail to ensure safe crossing of roundabouts by not considering the presence of third-party vehicles, leading to potential collision risks.
Innovation Solution
A method for selecting a traffic lane in a roundabout that takes into account the occupation of lanes by third-party vehicles and calculates a cost function to determine the safest lane change, using geolocation and telemetry data to assess collision risks and optimize lane selection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the navigation system determines the optimal lane based only on geolocation data and roundabout structure, then the lane selection is simple and fast, but the system cannot guarantee safe crossing by considering third-party vehicles
Solution Approach 1:
The lane selection process is divided into separate evaluation dimensions: geometric criteria (number of lanes to cross, position in roundabout) and dynamic safety criteria (occupancy of target lane by third-party vehicles). The cost function segments these factors into distinct components that can be independently calculated and then combined, allowing comprehensive safety assessment without overwhelming system complexity.
Solution Approach 2:
The system performs preliminary detection of third-party vehicles and calculation of lane occupancy probabilities before final lane selection. By pre-assessing the dynamic environment and computing cost values for each candidate lane, the system prepares safety information in advance, enabling reliable decision-making while maintaining real-time responsiveness.
2Reliability
If the system calculates cost function considering occupancy of all lanes by third-party vehicles, then collision risk is reduced, but computational load increases
Solution Approach 1:
The system focuses computational resources on evaluating specific candidate lanes that are geometrically suitable for reaching the desired exit, rather than uniformly analyzing all possible lanes. The cost function is selectively applied to a subset of relevant lanes identified through geometric pre-filtering, reducing overall computational load while maintaining comprehensive safety assessment for the most promising options.
Solution Approach 2:
The system transforms the complex problem of multi-vehicle collision risk assessment into a simplified cost function that aggregates occupancy probabilities across lanes. By changing the representation from detailed individual vehicle trajectories to aggregated lane occupancy parameters, the system reduces computational complexity while preserving essential safety information.
3Adaptability or versatility
If the motor vehicle frequently changes lanes to optimize for each exit, then the number of maneuvers increases, but the ability to adapt to different destinations improves
Solution Approach 1:
The system determines the optimal lane selection in advance based on the desired exit and current roundabout position, before the vehicle actually needs to execute the lane change. By pre-calculating the best lane for each possible destination, the system minimizes unnecessary maneuvers and allows the vehicle to commit to a lane early, reducing the frequency and time of lane changes while maintaining adaptability to different exits.
Data Source
Figure 1~3
Figure 4~5
Figure 6~7
AI summary
The invention concerns a method for selecting, for a motor vehicle (100) driving on a roundabout (1), a traffic lane (210, 220, 230) of a roundabout comprising several traffic lanes and a plurality of exits. According to the invention, the method comprises steps of: - detecting traffic lanes on the roundabout, - determining, for at least a portion of the traffic lanes, a first piece of data relating to the occupancy of the lane in question by third party vehicles, and a second piece of data relative to the number of traffic lanes to be crossed in order to change traffic lanes from a current traffic lane to the lane in question, - calculating the value of a cost function for each of the traffic lanes of the portion depending on the first piece of data and the second piece of data, and - selecting one of the traffic lanes for the motor vehicle depending on each calculated value of the cost function.