Intersection vehicle state detection and V2X (Vehicle-to-Everything) broadcasting method

Abstract

The invention relates to an intersection vehicle state detection and V2X (Vehicle-to-Everything) broadcasting method and belongs to the intelligent transportation technical field. The method includesthe following steps that: a system platform is set up, and then, initialization and image calibration are performed on the system; a license plate number identification and tracking-combined method isadopted to process images collected by each camera, so that the license plate number, coordinates, speed and time of a vehicle in a single camera image are obtained; the information of the i-th frameof a camera n is updated in an intersection vehicle global information list, and the Kalman filtering algorithm is adopted to estimate and update the states of vehicles in other cameras, and therefore, the license plate numbers, coordinates, speeds and time of all the vehicles at the intersection at the same time point are obtained; and V2X roadside equipment is adopted to broadcast vehicle global information to the vehicles within the range of the intersection. With the method of the invention adopted, intersection vehicle state detection and V2X broadcasting under a visual condition can berealized, and data support is provided for the passage decision making of the smart vehicles of the intersection.

Description

technical field [0001] The invention belongs to the technical field of intelligent transportation, and particularly relates to a vehicle state detection and V2X broadcasting method at an intersection. Background technique [0002] V2X (Vehicle-to-Everything) refers to the exchange of information between vehicles and the outside world, including V2V (Vehicle-to-Vehicle, vehicle-to-vehicle communication), V2I (Vehicle-to-Instruction, vehicle-to-road communication), V2P (Vehicle-to- Pedestrian, Car-Pedestrian Communication), etc. [0003] Self-driving cars have become the development trend of the current automobile industry. The existing environmental perception technology of self-driving cars can only realize environmental perception in certain scenarios, such as highways, closed parks, etc. However, for complex traffic scenarios, especially urban intersections, currently only relying on the self-driving car's own environmental perception system cannot accurately perceive the...

AI Technical Summary

Problems solved by technology

[0005] Among the existing vision-based vehicle state detection methods at intersections, some methods focus on improving the accuracy of vehicle state detection, and use relatively complex image recognition algorithms, but reduce the real-time detection while improving the detection accuracy. Some methods use the license plate recognition method for each frame image, which has a large amount of calculation and is not real-time; some methods track vehicle trajectories. There is a lack of one-to-one correspondence between license plate numbers and vehicle trajectories; and some methods only use a single camera to detect the vehicle state, which makes it impossible to detect the global vehicle state within the intersection range

[0006] In the existing multi-camera information splicing methods at intersections, some methods do not consider the vehicle state detection error caused by the asynchronous acquisition time of each camera when splicing the calculation results of multiple cameras; some vehicle detection methods use customized Surround view products with high-end cameras can avoid image splicing problems, but the price is relatively expensive

[0007] In the existing broadcasting and applications based on vehicle status at intersections, some apply the vehicle status at intersections to traffic warning, but some methods do not use trajectory prediction methods to solve the problem between different cameras when splicing information from different cameras. The problem of time asynchronousness; some only provide early warning on the roadside, and do not broadcast the vehicle status and early warning results to the corresponding vehicle or the driver on the vehicle through the V2X device

[0008] To sum up, the existing methods have their own advantages and disadvantages, but none of the methods has the following five characteristics at the same time: (1) lower cost; (2) better real-time performance; (3) higher accuracy ; (4) Detect the global vehicle status at the intersection; (5) Broadcast the global vehicle status to the intersection vehicle or vehicle driver through V2X

Images