Autonomous Vehicle Lateral Gap Planning Under Uncertainty
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in determining safe lateral gaps between themselves and other objects, particularly in dense environments, where large gaps may limit progress and cause vehicles to become stranded, due to uncertainties in measurements and predictions.
Innovation Solution
A method and system that calculate an uncertainty value by combining error values for an object's boundary, location, and predicted future location, adjusting the baseline lateral gap threshold based on driving conditions and object type, and attempting to find a trajectory that meets the threshold, with adjustments if necessary.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If large lateral gap thresholds are set to avoid potential collisions, then safety is improved, but vehicle progress is limited and vehicles may become stranded
Solution Approach 1:
The patent applies dynamics by making the lateral gap threshold adjustable rather than fixed. The system dynamically adapts the threshold based on uncertainty values calculated from sensor measurements and predictions. When uncertainty is low (precise measurements), smaller thresholds are used to improve progress. When uncertainty is high (imprecise measurements), larger thresholds are used to maintain safety. This resolves the contradiction by allowing the threshold to change based on real-time conditions.
Solution Approach 2:
The patent changes the parameter of lateral gap threshold based on uncertainty values. The uncertainty value is calculated from error values in boundary detection, location determination, and future location prediction. By changing the threshold parameter dynamically according to these uncertainty calculations, the system achieves both safety (when thresholds are larger) and progress (when thresholds are smaller), resolving the technical contradiction.
2Productivity
If lateral gap thresholds are reduced to improve progress in dense environments, then vehicle speed and progress are improved, but safety margins are reduced
Solution Approach 1:
The system dynamically adjusts the lateral gap threshold based on calculated uncertainty values. When the autonomous vehicle operates in conditions with high measurement precision (low uncertainty), the system can safely reduce the threshold to improve speed and progress. When measurement precision decreases (high uncertainty), the threshold increases to maintain safety margins. This dynamic adaptation resolves the contradiction between speed and safety.
Solution Approach 2:
The patent implements feedback by continuously calculating uncertainty values from sensor measurements and using these values to adjust the lateral gap threshold. The system monitors the precision of boundary detection, location determination, and prediction accuracy, then feeds this information back to adjust the safety threshold accordingly. This closed-loop feedback mechanism ensures that speed improvements do not compromise safety.
3Reliability
If uncertainty in measurements and predictions is high, then lateral gap thresholds must be increased for safety, but this limits operational flexibility in dense environments
Solution Approach 1:
The patent changes the lateral gap threshold parameter based on uncertainty levels. When uncertainty is high, the threshold increases to maintain safety. When uncertainty is low, the threshold decreases to improve operational flexibility. This parameter adaptation allows the system to operate flexibly in dense environments when measurements are precise while maintaining safety when measurements are uncertain.
Solution Approach 2:
The system dynamically adapts the lateral gap threshold to match current measurement and prediction quality. This dynamic behavior provides operational flexibility by allowing smaller thresholds when conditions permit (improving progress in dense environments) while automatically increasing thresholds when safety concerns arise from high uncertainty, thus resolving the contradiction between safety and flexibility.
Data Source
AI summary
Aspects of the disclosure provide for controlling an autonomous vehicle. For instance, a trajectory for the autonomous vehicle to traverse in order to follow a route to a destination may be generated. A first error value for a boundary of an object, a second error value for a location of the autonomous vehicle, a third error value for a predicted future location of the object may be received. An uncertainty value for the object may be determined by combining the first error value, the second error value, and the third error value. A lateral gap threshold for the object may be determined based on the uncertainty value. The autonomous vehicle may be controlled in an autonomous driving mode based on the lateral gap threshold for the object.