Parking lot parking space detection method based on multi-sensor information fusion

Abstract

The invention discloses a parking lot parking space detection method based on multi-sensor information fusion. The method comprises the following steps: step 1, placing a laser radar, a gyroscope and an accelerometer on a movement platform, and driving the movement platform to move in a parking lot; step 2, during a movement platform moving process, acquiring detection data of the laser radar and preprocessing the detection data; step 3, utilizing the preprocessed detection data for sequentially carrying out feature extraction and pattern matching, and recognizing local parking space information in the parking lot; step 4, analyzing data acquired by the gyroscope and the accelerometer, determining a position and a gesture of the movement platform, and correcting the local parking space information through gesture calculation of the movement platform; (5) overlaying multiple local environments to generate a scene map of parking lot empty parking spaces. The parking lot parking space detection method provided by the invention can be applicable to intelligent identification of parking spaces of an indoor parking lot, an outdoor parking lot, a roadside parking lot and the like.

Description

technical field [0001] The invention relates to a parking space detection method in a parking lot, in particular to a parking space detection method based on multi-sensor information fusion, which is applicable to parking spaces in indoor parking lots, outdoor parking lots and roadside parking lots, etc. Intelligent Recognition. Background technique [0002] In recent years, with the continuous increase of urban vehicles, sometimes it is difficult for drivers to park or find a car in the parking lot, which greatly reduces the service quality of the parking lot. At the same time, the use and maintenance of the existing parking lot management system, as well as personnel management costs are high, and the stability of the system needs to be strengthened. How to efficiently use the existing parking spaces has become a problem that must be solved in urban development. [0003] By consulting relevant patents and papers, it is found that the existing parking space management main...

AI Technical Summary

Problems solved by technology

[0004] Vehicles are detected by laying ground sense coils; the ground sense coil detection method is difficult to construct and needs to dig out the road surface, which is easy to cause damage to the road surface, and the coil is easily damaged and difficult to maintain;

[0005] Ultrasonic, infrared and other sensing methods are used to detect empty parking spaces; ultrasonic, infrared and other detection methods are affected by changes in weather, environment, temperature and airflow, and are not suitable for outdoor parking lots and all-weather use, and it is easy for non-vehicle objects to pass through the detection area Interference will cause inaccurate test results;

[0006] The camera-based visual detection method is to analyze and process the collected images; the video vehicle detection method is greatly affected by the lighting environment, and the detection accuracy is related to the placement of the camera. It is a static detection and cannot be flexibly maneuvered. At the same time, due to the need A certain number of cameras covering the parking lot, so the cost is higher;

[0007] Real-time detection of parking spaces based on ZigBee wireless sensor network and modulated infrared detection technology; based on ZigBee wireless sensor network and modulated infrared detection technology, it solves the influence of temperature and weather, and uses wireless communication to avoid damage to the ground, but It still needs to install sensors in each parking space, increase the investment in fixed assets of the parking lot, and due to the interference of various factors in the environment on signal detection and transmission, it will undoubtedly increase technical difficulty and increase maintenance costs.

Images