Autonomous Vehicle Perception Assistant Blind Spot Resolution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous driving vehicles (ADVs) have limited perception capabilities, particularly in detecting unusual road conditions and obstacles, especially at traffic junctions, due to their sensor systems' limitations, which can lead to missed detections and navigation challenges.
Innovation Solution
A perception assistant system that uses a network of sensors and image capturing devices to identify blind spots and obstacles, receiving images from remote servers to augment the ADV's perception, allowing it to generate trajectories to avoid obstacles and ensure safe navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the ADV relies only on its onboard sensor system, then the device complexity is reduced, but the perception capability and detection accuracy deteriorate due to limited detection distance and blind spots
Solution Approach 1:
The patent introduces image capturing devices as intermediary elements positioned at strategic locations (such as overhead cameras at intersections) to capture images that supplement the ADV's onboard sensor data. These intermediaries provide critical visual information from blind spots and distant locations without requiring the ADV to carry additional complex sensors, thus improving perception capability while maintaining reasonable device complexity
Solution Approach 2:
The patent transitions from relying solely on the ADV's ground-level sensor perspective to incorporating aerial or elevated imaging dimensions. By receiving images from overhead cameras or elevated positions, the system gains a top-down view that reveals blind spots and obstacles invisible to conventional onboard sensors, significantly enhancing detection accuracy without proportionally increasing device complexity
2Reliability
If the ADV monitors traffic from different directions at junctions, then the safety is improved, but the detection accuracy deteriorates because obstacles block critical perceptions from certain angles
Solution Approach 1:
The patent merges multiple information sources by combining data from the ADV's onboard sensors with images received from external image capturing devices. This fusion of ground-level sensor data and aerial/elevated camera images creates a comprehensive view that overcomes the limitations of any single source, improving both safety and detection accuracy by eliminating blind spots caused by obstacles
Solution Approach 2:
The system uses remotely positioned image capturing devices as intermediaries to provide unobstructed views of traffic conditions. These external cameras, positioned at elevated locations or strategic vantage points, capture images that bypass obstacles blocking the ADV's direct line of sight, thereby maintaining high detection accuracy while ensuring safe navigation through complex junctions
3Device complexity
If the ADV uses current sensor technologies, then the device complexity is maintained at acceptable levels, but the detection capability deteriorates by missing unusual road conditions and temporary construction zones
Solution Approach 1:
The patent employs image capturing devices positioned at strategic locations along the road network as intermediaries to detect and report unusual road conditions, temporary construction zones, and other environmental features. These external observers provide critical information about road conditions without requiring the ADV to carry specialized sensors, maintaining acceptable device complexity while significantly improving detection capability
Solution Approach 2:
The system performs preliminary detection of road conditions and obstacles by having image capturing devices continuously monitor and transmit images of road conditions, construction zones, and potential hazards before the ADV reaches those locations. This advance detection allows the navigation system to plan routes that avoid problematic areas, improving detection capability without adding complex real-time sensing equipment to the vehicle
Data Source
AI summary
In one embodiment, a system of an ADV perceives a driving environment surrounding the ADV using a plurality of sensors mounted on the ADV. The system identifies a blind spot based on the perceived driving environment surrounding the ADV. The system in response to identifying the blind spot, receives an image having the blind spot from an image capturing device disposed within a predetermined proximity of the blind spot. In some embodiments, the system receives the image having the blind spot from a remote server communicatively coupled to the image capturing device. The system identifies an obstacle of interest at the blind spot of the ADV based on the image. The system generates a trajectory based on the obstacle of interest at the blind spot to control the ADV to avoid the obstacle of interest.