Side-View Camera Cross-Traffic Detection for Occluded Intersections
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous driving systems face challenges in accurately navigating various road conditions and avoiding collisions with moving and stationary objects, particularly at intersections, due to limitations in sensor range and occlusions, as well as high costs associated with using LiDAR and RADAR sensors for cross-road angle configurations.
Innovation Solution
A camera vision method using side-view cameras to process road images, generate bounding boxes for objects, determine their direction and speed, and send control signals to the autonomous vehicle to ensure safe movement, reducing the need for extensive LiDAR or RADAR sensors by employing machine learning and dynamic image cropping for long-distance detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If LiDAR and RADAR sensors are used for cross-road angle configurations, then measurement precision and detection reliability are improved, but device complexity and cost increase
Solution Approach 1:
The patent employs a single camera sensor to perform multiple functions: detecting objects, determining their direction of travel, calculating speeds, and assessing safety for autonomous vehicle movement. This multi-functional approach replaces the need for multiple specialized sensors (LiDAR, RADAR) while achieving comparable detection capabilities through computer vision algorithms and machine learning models.
2Reliability
If LiDAR and RADAR sensors are used for cross-road angle configurations, then measurement precision and detection reliability are improved, but cost increases
Solution Approach 1:
The patent substitutes expensive, complex LiDAR and RADAR sensors with relatively inexpensive camera sensors. While individual cameras are less costly than LiDAR/RADAR systems, the solution compensates for the lower per-sensor cost through sophisticated software processing and machine learning algorithms that extract maximum information from the camera data, achieving reliable detection at lower overall system cost.
3Device complexity
If camera-based systems are used for object detection, then device complexity and cost are reduced, but measurement precision and detection range are limited
Solution Approach 1:
The system captures a series of images over time before making safety determinations, performing preliminary detection and tracking of objects. By analyzing object positions across multiple frames, the system can calculate direction of travel and speed, thereby compensating for the limited precision of individual camera frames and achieving more accurate measurement through temporal accumulation of data.
Solution Approach 2:
The patent transitions from spatial analysis alone to spatio-temporal analysis by incorporating the time dimension. Instead of relying solely on single-frame image precision, the system analyzes object movement across multiple time points, extracting velocity and trajectory information that enhances measurement precision and extends effective detection range through motion-based detection.
4Measurement precision
If computational algorithms process all road images to determine object parameters, then measurement precision is improved, but productivity and processing speed decrease
Solution Approach 1:
The patent divides the road scene into multiple regions of interest, particularly focusing on cross-traffic areas and potential collision zones. By segmenting the image processing task and applying specialized algorithms only to relevant regions rather than processing entire images uniformly, the system maintains high measurement precision for critical parameters while improving overall processing speed and productivity.
Data Source
AI summary
Autonomous vehicles must accommodate various road configurations such as straight roads, curved roads, controlled intersections, uncontrolled intersections, and many others. Autonomous driving systems must make decisions about the speed and distance of traffic and about obstacles including obstacles that obstruct the view of the autonomous vehicle's sensors. For example, at intersections, the autonomous driving system must identify vehicles in the path of the autonomous vehicle or potentially in the path based on a planned path, estimate the distance to those vehicles, and estimate the speeds of those vehicles. Then, based on those and the road configuration and environmental conditions, the autonomous driving system must decide whether it is safe to proceed along the planned path or not, and when it is safe to proceed.