Autonomous Vehicle Encroachment Detection Using Expanded Bounding Boxes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional automated vehicles struggle to identify and react to traffic vehicles that are not conforming to driving norms, which can lead to safety issues due to the vehicles drifting too closely to the automated vehicle.
Innovation Solution
An autonomy system for automated vehicles that detects and avoids lazy or unpredictable vehicles by generating bounding boxes for tracking traffic vehicles, simulating their behavior, and determining if they are encroaching based on overlapping or entering the automated vehicle's lane.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional automated vehicles use standard obstacle detection methods, then they can identify clearly dangerous obstacles (e.g., stopped vehicles), but they fail to identify and react to problematic traffic vehicles that are drifting too closely or failing to conform with driving norms
Solution Approach 1:
The system performs preliminary action by generating expanded bounding boxes around close vehicles before actual encroachment occurs. This allows the system to predict potential lane violations and take preventive measures. The expanded bounding box extends into the automated vehicle's lane, enabling early detection of vehicles that are drifting too closely but have not yet crossed the lane line.
Solution Approach 2:
The system applies preliminary anti-action by simulating future positions of close vehicles and predicting their trajectories. This allows the automated vehicle to prepare avoidance maneuvers in advance, counteracting the potential harmful effect of encroachment before it actually occurs. The motion planning component generates possible avoidance maneuvers based on predicted encroachment scenarios.
2Measurement precision
If the autonomy system uses expanded bounding boxes to detect encroaching vehicles, then it can identify vehicles drifting too closely, but this requires additional processing complexity compared to standard bounding box detection
Solution Approach 1:
The system applies local quality by selectively generating expanded bounding boxes only for close vehicles that are within a threshold distance, rather than expanding all bounding boxes uniformly. This localized approach improves detection accuracy for potentially encroaching vehicles while minimizing unnecessary processing complexity for vehicles that are clearly at safe distances.
Solution Approach 2:
The system uses partial action by applying expanded bounding boxes only to close vehicles that meet specific criteria (within threshold distance), rather than applying the expansion to all detected vehicles. This selective approach provides sufficient detection capability for high-risk scenarios while reducing overall computational burden.
3Reliability
If the autonomy system simulates future positions of close vehicles to predict encroachment, then it can proactively avoid potential hazards, but this increases computational requirements and processing time
Solution Approach 1:
The system performs preliminary simulation of vehicle trajectories only for close vehicles that are identified as potential risks. By pre-computing predicted positions and potential encroachment scenarios for high-risk vehicles, the system enables proactive safety measures while avoiding unnecessary computational energy consumption for vehicles that are clearly at safe distances.
Solution Approach 2:
The trajectory simulation and predictive analysis are applied locally only to close vehicles within the threshold distance, rather than performing comprehensive simulation on all detected vehicles. This localized predictive approach maintains high reliability for potential hazards while minimizing overall computational energy requirements.
Data Source
AI summary
Embodiments herein include an autonomy system of an automated vehicle performing operations for detecting and avoiding lazy or unpredictable vehicles on a roadway. The autonomy system generates bounding boxes for tracking traffic vehicles recognized in sensor data. For close vehicles, the autonomy system generates expanded bounding boxes. The autonomy system determines the close vehicle is an encroaching vehicle by detecting that the expanded bounding box of the close vehicle overlapped the middle lane line or enters into the automated vehicle's current lane of travel. The autonomy system may simulate the behavior of the close vehicle by forward-propagating predicted positions of the close vehicle and expanded bounding box over some future time. In response to detecting the encroaching vehicle, the autonomy system may determine whether to perform an avoidance action or select a particular avoidance action, such as biasing away from the encroaching vehicle.