UAV-Assisted Autonomous Driving for Occlusion-Aware Perception
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous driving systems face challenges in detecting objects blocked by other vehicles and have limited range, especially in low visibility weather conditions, as they rely on sensors like LiDAR and cameras that struggle to perceive objects beyond a certain distance or behind obstacles.
Innovation Solution
An autonomous driving system that incorporates an unmanned aerial vehicle (UAV) to collect and transmit ground traffic information, which is combined with data from land vehicle sensors to generate a comprehensive world model, enhancing perception and decision-making capabilities by overcoming blind spots and expanding detection range.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If LiDAR, cameras, or radar sensors are used for object detection, then the system can detect objects within a limited range (around one hundred meters), but it cannot detect objects blocked by other vehicles or in low visibility weather conditions
Solution Approach 1:
The patent introduces aerial surveillance from a third dimension (air space) to complement traditional ground-based two-dimensional sensor arrays. UAVs equipped with cameras and sensors operate from elevated positions, providing top-down views that penetrate occlusions and extend the detection envelope vertically and horizontally beyond the vehicle's immediate vicinity.
Solution Approach 2:
The system employs communication modules and data fusion algorithms as intermediaries to bridge the gap between UAV-collected aerial data and the vehicle's decision-making system. The communication module transmits processed information from the UAV to the vehicle, enabling the ground vehicle to benefit from extended-range detection without direct physical contact between sensors and distant objects.
2Measurement precision
If multiple sensors are deployed to improve detection capability, then the system can detect objects within its range, but the device complexity increases
Solution Approach 1:
The system divides the detection task into two segments: ground-based sensors handle immediate vicinity detection, while aerial UAVs handle extended-range and occlusion-penetration detection. This segmentation allows each subsystem to operate independently within its optimal range, reducing the complexity burden on any single sensor suite while achieving comprehensive coverage through coordinated operation.
Data Source
AI summary
Aspects an autonomous driving system with air support are described herein. The aspects may include an unmanned aerial vehicle (UAV) in the air and a land vehicle on the ground communicatively connected to the UAV. The UAV may include at least one UAV camera configured to collect first ground traffic information and a UAV communication module configured to transmit the collected first ground traffic information. The land vehicle may include one or more vehicle sensors configured to collect second ground traffic information surrounding the land vehicle, a land communication module configured to receive the first ground traffic information from the UAV, and a processor configured to combine the first ground traffic information and the second ground traffic information to generate a world model.