Autonomous Vehicle Occupancy Thresholds for Sensor Blind Spots
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles in confined areas face challenges with sensor coverage limitations, leading to potential deadlocks and inaccurate obstacle detection, especially when sensors' field of view is obstructed or insufficient, resulting in unsafe and inefficient operation.
Innovation Solution
A method that dynamically adjusts occupancy thresholds based on the vehicle's position and sensor field of view, allowing movement into areas with lower accuracy and reliability, while ensuring safety by disabling movement into areas with uncertain occupancy scores, thereby preventing deadlocks and improving route planning accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the occupancy threshold is set high to ensure safety in areas with uncertain occupancy scores, then safety is improved, but the vehicle cannot move into potentially free areas, leading to deadlocks and reduced productivity
Solution Approach 1:
The occupancy threshold is made dynamic rather than static. The threshold adapts based on the vehicle's current position and the specific area being evaluated, allowing the system to be conservative when needed and permissive when safe, thus resolving the contradiction between safety and productivity
Solution Approach 2:
Different occupancy thresholds are applied to different areas of the map based on local characteristics. Areas with high uncertainty receive higher thresholds for safety, while areas with sufficient sensor coverage receive lower thresholds to enable movement, preventing deadlocks while maintaining safety
2Productivity
If the occupancy threshold is set low to allow movement into areas with uncertain occupancy scores, then productivity is improved, but safety is compromised as the vehicle may move into occupied areas
Solution Approach 1:
The system dynamically adjusts the occupancy threshold based on the vehicle's position and sensor coverage characteristics, rather than using a fixed low threshold. This allows the vehicle to move efficiently into areas with sufficient information while maintaining high safety standards in areas with uncertain occupancy scores
3Reliability
If sensors are added to extend the total field of view to avoid deadlocks, then the deadlock problem is solved, but device complexity and cost increase
Solution Approach 1:
Instead of adding hardware sensors, the system changes the parameter of the occupancy threshold dynamically based on sensor coverage and position. This software-based solution achieves deadlock prevention without increasing device complexity or cost
Solution Approach 2:
The occupancy threshold acts as an intermediary mechanism that mediates between the vehicle's movement desires and the sensor coverage limitations. It allows the system to handle incomplete sensor information intelligently without requiring additional sensors
Data Source
AI summary
A method of controlling an autonomous vehicle, which is movable on a surface, includes obtaining a model (world model) of the surface, by which each area of the surface is associated with a probabilistic occupancy score; determining, on the basis of the area's position, an occupancy threshold to be applied to an area of the surface; enabling movement of the AV into the area if the associated occupancy score is less than the determined occupancy threshold; and otherwise disabling movement into the area. In one embodiment, where the model is obtained or updated based on measurement data from one or more sensors carried by the autonomous vehicle, the occupancy threshold is determined to be relatively lower if the area is outside a field of view of the sensors carried by the AV and relatively higher if the area is inside the field of view.