Underground parking lot vehicle positioning system and method

Abstract

The invention discloses an underground parking lot vehicle positioning system and method, and the system combines a VLC positioning technology with a Bayesian probability model, and achieves the precise positioning of a vehicle in a complex underground parking lot environment. By introducing an empty parking space identification module and a display unit, a vehicle is guided to quickly and accurately find an empty parking space in an underground parking lot and park the empty parking space; parking space number information and license plate number information recognized by a parking space number recognition device and a front-end vehicle recognition device are combined, the matched parking space number information and license plate number information are checked through a mobile phone, and a driver is assisted to quickly find a vehicle parked on an empty parking space. The system realizes accurate vehicle positioning and auxiliary vehicle searching functions of the underground parking lot, is simple in design and relatively low in cost, and has the characteristics of high precision, good real-time performance and reliability and the like.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle positioning, and in particular to a vehicle positioning system and method for an underground parking lot. Background Art

[0002] In recent years, with the continuous growth of the number of vehicles, underground parking lots have been widely used in order to effectively utilize space and ensure vehicle safety. However, due to the large scale and complex environment of underground parking lots, current technical means face challenges in obtaining accurate vehicle location information and vacant parking space information, which directly leads to the increasingly prominent problems of difficulty in finding and parking in underground parking lots.

[0003] Due to the obstruction of building walls, traditional GPS and Beidou satellite systems cannot be directly used for positioning in underground parking lots. At present, the methods for achieving underground parking lot positioning mainly include UWB (ultra-wideband) positioning system, magnetic nail positioning system, monocular or binocular visual positioning system; UWB positioning system obtains the positioning information of underground parking lots by deploying a large number of base stations, but the overall technical cost is relatively expensive and the adaptability to the environment is weak; the magnetic nail positioning system obtains accurate positioning information of underground parking lots by burying magnetic nails underground through the transformation of parking lots, but the deployment process is relatively cumbersome and the underground parking lots must be transformed, which is time-consuming and labor-intensive; the monocular or binocular visual positioning system mainly relies on cameras to capture images to obtain the positioning information of underground parking lots, but the underground parking lot environment is complex and the light is easily limited, resulting in inaccurate positioning.

[0004] VLC positioning technology is mainly based on LED lamps as signal transmitters. The receiving device receives and analyzes the light signals emitted by the LED to achieve positioning. These light signals are usually encoded into specific patterns or sequences to represent different location information. The receiving device uses photosensitive elements to capture these light signals and restores the original location information through decoding algorithms. However, some encoding algorithms, such as the two-step weighted least squares algorithm, are greatly affected by changes in background light intensity and cannot adapt well to the complex light environment in underground parking lots.

[0005] The patent application with application number CN202110170044.9 proposes a UWB-based parking lot vehicle positioning method and system, which includes a positioning tag, a UWB positioning base station and a background server. The positioning tag is used to receive data packets sent by the UWB positioning base station, perform position calculation, and send the calculated data to the background server through the network; the UWB positioning base station is used to send positioning data packets to the UWB positioning tag; the background server sends the location data to the user APP and the parking lot management client via the Internet, and the background server can also plan the navigation path and send it to the user APP; the positioning system adopts a passive TDOA calculation method, which greatly reduces the operating pressure of the server; however, UWB signals are easily blocked, reflected and scattered by obstacles such as pillars and walls in underground parking lots, resulting in multipath effects, resulting in signal attenuation and reduced positioning accuracy. Summary of the invention

[0006] In order to overcome the shortcomings of the above-mentioned prior art, the purpose of the present invention is to provide a vehicle positioning system and method for an underground parking lot. The system combines VLC positioning technology with a Bayesian probability model to achieve precise positioning of vehicles in a complex underground parking lot environment; by introducing an empty parking space recognition module and a display unit, the vehicle is guided to quickly and accurately find an empty parking space in the underground parking lot and park; the parking space number information and license plate number information identified by a parking space number recognition device and a front-end vehicle recognition device are combined, and the matching parking space number information and license plate number information are viewed through a mobile phone to assist the driver in quickly finding the vehicle parked in the empty parking space; the system realizes the precise positioning of vehicles in the underground parking lot and the auxiliary vehicle search function, has a simple design, and is relatively low in cost, and at the same time has the characteristics of high precision, good real-time performance and reliability.

[0007] In order to achieve the above object, the technical solution adopted by the present invention is:

[0008] An underground parking lot vehicle positioning system, comprising a front-end vehicle identification device, a VLC positioning device, an empty parking space identification module, a display unit, a parking space number identification device, a mobile phone, and a wireless transmission unit;

[0009] The front-end vehicle identification device is used to detect the license plate number information of the vehicle entering the underground parking lot, and transmit the license plate number information directly to the parking space number identification device and indirectly synchronize it to the mobile phone through the wireless transmission unit;

[0010] The VLC positioning device is used to detect and locate the position information of the vehicle entering the underground parking lot in real time, and transmit the real-time position information of the vehicle to the display unit through the wireless transmission unit;

[0011] The vacant parking space identification module is used to detect the vacant parking space information in the underground parking lot in real time, and transmit the vacant parking space information in the underground parking lot to the display unit through the wireless transmission unit;

[0012] The display unit is used to visualize the real-time position information of the vehicle transmitted by the VLC positioning device and the information of the available parking spaces in the underground parking lot transmitted by the empty parking space recognition module, and guide the driver to park the vehicle in the empty parking space;

[0013] The parking space number recognition device is used to recognize the parking space number information corresponding to the vehicle parked in the empty parking space, and receive the license plate number information transmitted by the front-end vehicle recognition device, and match the parking space number information with the license plate number information and transmit it to the mobile phone through the wireless transmission unit;

[0014] The mobile phone is used to receive the license plate number information transmitted by the front-end vehicle identification device, the matching parking space number information and the license plate number information transmitted by the parking space number identification device, and assist the driver in finding an empty parking space in the underground parking lot, thereby finding the vehicle parked in the empty parking space;

[0015] The wireless transmission unit adopts wireless communication to assist the front-end vehicle identification device, the VLC positioning device, the empty parking space identification module, the display unit, the parking space number identification device, and the information transmission between the mobile phone.

[0016] Furthermore, the front-end vehicle identification device includes a first camera, an information processing unit, and an information transmission unit;

[0017] The first camera is used to collect images of vehicles entering the underground parking lot and transmit the collected vehicle images to the information processing unit;

[0018] The information processing unit recognizes and processes the vehicle image captured by the first camera, thereby obtaining the vehicle license plate number information;

[0019] After receiving the vehicle license plate number information acquired by the information processing unit, the information transmission unit directly transmits the license plate number information to the parking space number transmission unit of the parking space number recognition device and indirectly synchronizes it to the mobile phone through the wireless transmission unit.

[0020] Further, the VLC positioning device includes a VLC transmitter, a VLC receiver, and a VLC positioning module;

[0021] The VLC transmitter includes a plurality of LED lamps and a digital signal processor; the digital signal processor sequentially pre-processes, encodes, and modulates the pulse signal, and uses the modulated pulse signal to control the LED lamp to emit visible light loaded with the lamp position information and ID information;

[0022] The VLC receiver includes a photoelectric converter and a signal processing module; after receiving the visible light loaded with the location information and ID information of the lamp, the photoelectric converter converts the optical signal into an electrical signal; the signal processing module amplifies and filters the received electrical signal, and then uses a demodulation algorithm to convert the amplified and filtered electrical signal into a digital signal, and uses a mean filter to calculate the RSSI value corresponding to the received digital signal; finally, the RSSI information is transmitted to the VLC positioning module by wired transmission;

[0023] The VLC positioning module is used to receive the RSSI value of the VLC receiver, establish a conversion relationship model between RSSI and distance, and estimate the distance between the VLC receiver and each VLC transmitter; based on the estimated distances between more than three VLC receivers and each VLC transmitter and the known positions of the VLC transmitters, the real-time position information of the vehicle is calculated using a Bayesian probability model; the VLC positioning module transmits the real-time position information of the vehicle to a display unit through a wireless transmission unit.

[0024] Furthermore, the use of the Bayesian probability model to calculate the real-time position information of the vehicle specifically includes: assuming that the initial position of the vehicle obeys a Gaussian distribution, at each time step, predicting the position of the vehicle at the next moment based on the vehicle motion model and the historical trajectory; updating the prior probability based on the predicted position of the vehicle at the next moment; at each time step, calculating the likelihood probability based on the RSSI value received by the VLC receiver; combining the calculated prior probability and likelihood probability, updating the posterior probability distribution, and normalizing the posterior probability distribution, and estimating the real-time position of the vehicle through the normalized posterior probability distribution.

[0025] Further, the empty parking space identification module includes a parking lot map unit, a parking lot space unit, a geomagnetic vehicle detector, a parking space status unit, a parking space information processing unit, and a parking space information transmission module;

[0026] The parking lot map unit stores map information of the underground parking lot where the vehicle is located;

[0027] The parking lot parking space unit stores the location information of all parking spaces in the underground parking lot;

[0028] The geomagnetic vehicle detector is used to detect whether there is a vehicle parked in the parking space;

[0029] The vehicle status unit is used to receive the parking status of the vehicle detected by the geomagnetic vehicle detector, and mark the occupancy status of the parking space according to the parking status of the vehicle detected by the geomagnetic vehicle detector;

[0030] The parking space information processing unit determines whether there are vacant parking spaces in the underground parking lot, determines the positions of the vacant parking spaces, and calculates the number of vacant parking spaces based on the map information of the underground parking lot where the vehicle is located stored in the parking lot map unit, the position information of all parking spaces in the underground parking lot stored in the parking lot parking space unit, and the parking space occupancy status transmitted by the parking space status unit; the parking space information processing unit transmits the position information of the vacant parking spaces and the number information of the vacant parking spaces to the parking space information transmission module by wired transmission;

[0031] The parking space information transmission module is used to receive the position information of the empty parking spaces and the number information of the empty parking spaces from the parking space information processing unit, and transmit the information to the display unit through the wireless transmission unit.

[0032] Furthermore, the display unit includes an on-board display screen for visualizing the real-time position information of the vehicle transmitted by the VLC positioning module and the position information and number information of empty parking spaces transmitted by the parking space information transmission module, so as to guide the driver to park the vehicle in an empty parking space.

[0033] Further, the parking space number recognition device includes a second camera, an image processing unit and a parking space number transmission unit;

[0034] The second camera is used to collect the rear environment information of the vehicle in the parking space, and transmit the collected rear environment information of the vehicle to the image processing unit;

[0035] The image processing unit is used to receive the vehicle rear environment information transmitted by the second camera, and classify and extract the vehicle rear environment information through the KL dimension reduction kernel Fisher judgment parking space detection algorithm, so as to identify the parking space number information; firstly, the vehicle rear environment information is subjected to KL dimension reduction, so that the information in the vehicle rear environment information that is conducive to classification is concentrated in a small number of dimensions; secondly, the kernel Fisher discriminant analysis is used to perform parking space parking discrimination, and the parking space feature parameters are mapped to a new feature space through the kernel function, and then the Fisher linear discriminant analysis is performed, so as to identify the parking space number information and transmit it to the parking space number transmission unit;

[0036] The parking space number transmission unit is used to receive the parking space number information transmitted by the image processing unit and the license plate number information transmitted by the information transmission unit, and then match the parking space number information with the license plate number information to the driver's mobile phone through the wireless transmission unit;

[0037] A method for locating vehicle information in an underground parking lot, comprising the following steps:

[0038] S1. When a vehicle enters an underground parking lot, the front-end vehicle identification device detects the license plate number information, and transmits the license plate number information directly to the parking space number identification device and indirectly synchronizes it to the mobile phone through the wireless transmission unit;

[0039] S2. After the vehicle enters the underground parking lot, the VLC positioning device detects and locates the real-time position information of the vehicle, and transmits the real-time position information of the vehicle to the display unit through the wireless transmission unit;

[0040] S3, the vacant parking space recognition module detects the vacant parking space information in the underground parking lot, and transmits the vacant parking space information in the underground parking lot to the display unit through the wireless transmission unit;

[0041] S4, the display unit visualizes the real-time position information of the vehicle transmitted by the VLC positioning device in step S2 and the real-time vacant parking space information transmitted by the vacant parking space recognition module in step S3 on the vehicle display screen, guiding the driver to park the vehicle in the designated vacant parking space;

[0042] S5. After the driver parks the vehicle in an empty parking space, the parking space number recognition device recognizes the parking space number information corresponding to the vehicle parked in the empty parking space, and combines the license plate number information transmitted by the front-end vehicle recognition device in step S1 to match the parking space number information and the license plate number information, and then transmits them to the mobile phone through the wireless transmission unit; the driver searches for an empty parking space based on the matching parking space number information and license plate number information on the mobile phone, thereby finding the vehicle parked in the empty parking space.

[0043] Furthermore, the step S2 is specifically as follows:

[0044] Step S2-1: The VLC transmitter of the VLC positioning device emits visible light loaded with the lamp location information and ID information;

[0045] Step S2-2: The VLC receiver of the VLC positioning device receives the visible light loaded with the lamp location information and ID information in step S2-1, and first converts the optical signal into an electrical signal; then amplifies and filters the electrical signal, and uses a demodulation algorithm to convert the amplified and filtered electrical signal into a digital signal; finally, uses mean filtering to calculate the RSSI value corresponding to the received digital signal;

[0046] Step S2-3: After receiving the RSSI value in step S2-2, the VLC positioning module of the VLC positioning device establishes a conversion relationship model between RSSI and distance, and estimates the distance between the VLC receiver and each VLC transmitter; based on the estimated distances between more than three VLC receivers and each VLC transmitter and the known positions of the VLC transmitters, the real-time position information of the vehicle is calculated using a Bayesian probability model; the VLC positioning module transmits the real-time position information of the vehicle to the display unit through the wireless transmission unit.

[0047] Furthermore, the step S3 is specifically as follows:

[0048] Step S3-1: Acquire the map information of the underground parking lot stored in the parking lot map unit of the empty parking space recognition module;

[0049] Step S3-2: Acquire the location information of all parking spaces in the underground parking lot stored in the parking lot unit of the empty parking space identification module;

[0050] Step S3-3: obtaining information on vehicles parked in parking spaces detected by the geomagnetic vehicle detector of the empty parking space recognition module;

[0051] Step S3-4: the vehicle status unit of the empty parking space recognition module marks the occupancy status of the parking space according to the information of the vehicle parked in the parking space in step S3-3;

[0052] Step S3-5: the parking space information processing unit of the empty parking space identification module determines the location of the empty parking space and calculates the number of empty parking spaces according to the map information of the underground parking lot in step S3-1, the location information of all parking spaces in the underground parking lot in step S3-2, and the occupancy status of the parking spaces in step S3-4;

[0053] Step S3-6: After receiving the position information and number information of the empty parking spaces in step S3-5, the parking space information transmission module of the empty parking space identification module transmits the information to the display unit via the wireless transmission unit.

[0054] Compared with the prior art, the beneficial effects of the present invention are:

[0055] 1. The present invention realizes the accurate positioning of vehicles in underground parking lots by applying VLC positioning technology to underground parking lots, effectively solving the problem that traditional positioning technology is difficult to play a role due to multiple environmental factors such as dim light, complex structure, and signal shielding in underground parking lots; at the same time, the VLC positioning device can use the existing LED lighting equipment in the underground parking lot as a signal transmission source, thereby improving the utilization rate of equipment and energy.

[0056] 2. The present invention adopts the Bayesian probability model to effectively integrate multiple information sources, update the position probability distribution in real time according to the newly received signal, realize dynamic tracking and positioning, and continuously optimize the position estimation by constantly acquiring new signal features, thereby providing more accurate positioning information; at the same time, the Bayesian probability model has better adaptability in parking lots with complex light sources, can maintain a high positioning accuracy under different background light intensities, and has good anti-interference and robustness.

[0057] In summary, the present invention combines VLC positioning technology with the Bayesian probability model to achieve accurate positioning of vehicles in a complex underground parking environment; by introducing an empty parking space recognition module and a display unit, the vehicle is guided to quickly and accurately find an empty parking space in the underground parking lot and park; the parking space number information and license plate number information recognized by the parking space number recognition device and the front-end vehicle recognition device are combined, and the matching parking space number information and license plate number information are viewed through a mobile phone to assist the driver to quickly find the vehicle parked in the empty parking space; the system realizes the accurate positioning of vehicles in the underground parking lot and the auxiliary vehicle search function, with a simple design and relatively low cost, and at the same time has multiple advantages such as high precision, good real-time performance and reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 It is the overall structure diagram of the underground parking lot vehicle positioning system of the present invention.

[0059] Figure 2 The figure is a flow chart of the vehicle positioning method in the underground parking lot of the present invention.

[0060] Figure 3 The simulation results of estimating the vehicle's driving trajectory using the VLC positioning method under the least squares method and the VLC positioning method under the Bayesian probability model when the vehicle of the present invention passes through the underground parking lot at a constant speed under different background light intensities respectively; Figure 3 (a) shows the simulation results of estimating the vehicle's driving trajectory using the VLC positioning method under the least squares method and the VLC positioning method under the Bayesian probability model when the vehicle is traveling at a constant speed in an underground parking lot under a background light intensity of 300 lux. Figure 3 (b) shows the simulation results of estimating the vehicle's driving trajectory using the VLC positioning method under the least squares method and the VLC positioning method under the Bayesian probability model when the vehicle is traveling at a constant speed in an underground parking lot under a background light intensity of 400 lux. Figure 3 (c) shows the simulation results of estimating the vehicle's driving trajectory using the VLC positioning method under the least squares method and the VLC positioning method under the Bayesian probability model when the vehicle is traveling at a constant speed in an underground parking lot under a background light intensity of 500 lux. Figure 3 (d) in the figure is the simulation result of estimating the vehicle's driving trajectory using the VLC positioning method under the least squares method and the VLC positioning method under the Bayesian probability model when the vehicle is traveling at a constant speed in an underground parking lot under a background light intensity of 600 lux.

[0061] Figure 4 The graph is a curve showing the root mean square error variation of the vehicle position estimated by the VLC positioning method based on the least squares method and the VLC positioning method based on the Bayesian probability model when the vehicle of the present invention travels at a constant speed in an underground parking lot under a background light intensity of 500 lux. DETAILED DESCRIPTION

[0062] The technical solution of the present invention is described in detail below with reference to the accompanying drawings and embodiments.

[0063] See also Figure 1 , an underground parking lot vehicle positioning system, comprising a front-end vehicle identification device, a VLC positioning device, an empty parking space identification module, a display unit, a parking space number identification device, a mobile phone, and a wireless transmission unit;

[0064] The front-end vehicle identification device is installed at the entrance of the underground parking lot; the front-end vehicle identification device is used to detect the license plate number information of the vehicle entering the underground parking lot, and transmit the license plate number information directly to the parking space number identification device and indirectly synchronize it to the mobile phone through the wireless transmission unit;

[0065] The front-end vehicle identification device includes a first camera, an information processing unit, and an information transmission unit; the first camera is implemented based on TC-CJ01; the information processing unit is implemented based on a development board of an STM32 single-chip microcomputer; and the information transmission unit is implemented based on an HC-06 Bluetooth module.

[0066] The first camera is installed at the entrance gate of the underground parking lot, and is used to collect images of vehicles entering the underground parking lot, and transmit the collected vehicle images to the information processing unit; the information processing unit recognizes and processes the vehicle images collected by the first camera, so as to obtain the vehicle license plate number information; by performing preprocessing, feature extraction, character segmentation and character recognition on the vehicle images captured by the first camera in sequence, the vehicle license plate number is finally recognized.

[0067] After receiving the vehicle license plate number information acquired by the information processing unit, the information transmission unit directly transmits the license plate number information to the parking space number transmission unit of the parking space number recognition device and indirectly synchronizes it to the mobile phone through the wireless transmission unit.

[0068] The VLC positioning device is used to detect and locate the position information of the vehicle entering the underground parking lot in real time, and transmit the real-time position information of the vehicle to the display unit through the wireless transmission unit;

[0069] The VLC positioning device includes a VLC transmitter, a VLC receiver, and a VLC positioning module; the VLC transmitter is implemented based on the WICOP series high-power LED lamp of Seoul Semiconductor; the VLC receiver is implemented based on the S1223 of Hamamatsu; and the VLC positioning module is implemented based on the CC2431 of TI.

[0070] The VLC transmitter is installed on the top of the underground parking lot, and is arranged at an equal distance of 3 meters to 5 meters. The VLC transmitter includes multiple LED lamps and a digital signal processor; the digital signal processor pre-processes, encodes, and modulates the pulse signal in sequence, and uses the modulated pulse signal to control the LED lamp to emit visible light loaded with the lamp location information and ID information; in addition, the existing LED lighting equipment on the top of the underground parking lot can be used as a signal transmission source, thereby improving the utilization rate of equipment and energy.

[0071] The VLC receiver is installed on the top of the vehicle; the VLC receiver includes a photoelectric converter and a signal processing module; after the photoelectric converter receives the visible light loaded with the lamp position information and ID information, it converts the optical signal into an electrical signal; the signal processing module amplifies and filters the received electrical signal, and then uses a demodulation algorithm to convert the amplified and filtered electrical signal into a digital signal, and uses mean filtering to calculate the RSSI value corresponding to the received digital signal; finally, the RSSI information is transmitted to the VLC positioning module by wired transmission;

[0072] The VLC positioning module is used to receive the RSSI value of the VLC receiver, establish a conversion relationship model between RSSI and distance, and estimate the distance between the VLC receiver and each VLC transmitter; based on the estimated distances between more than three VLC receivers and each VLC transmitter and the known positions of the VLC transmitters, the real-time position information of the vehicle is calculated using a Bayesian probability model; the VLC positioning module transmits the real-time position information of the vehicle to a display unit through a wireless transmission unit.

[0073] The method of using the Bayesian probability model to calculate the real-time position information of the vehicle specifically includes: assuming that the initial position of the vehicle obeys a Gaussian distribution, predicting the position of the vehicle at the next moment according to the vehicle motion model and the historical trajectory at each time step; updating the prior probability according to the predicted position of the vehicle at the next moment; calculating the likelihood probability according to the RSSI value received by the VLC receiver at each time step; updating the posterior probability distribution based on the calculated prior probability and likelihood probability, normalizing the posterior probability distribution, and estimating the real-time position of the vehicle through the normalized posterior probability distribution.

[0074] According to the newly received signal, the position probability distribution is updated in real time to achieve dynamic tracking and positioning, continuously obtain new signal features, and continuously optimize the position estimate, rather than relying solely on the initial positioning parameters. The Bayesian probability model provides an effective multi-modal (such as light intensity, angle, etc.) positioning information fusion framework. Previous methods may simply superimpose or separately process different types of information, while it can organically combine this information and give full play to the advantages of various positioning features. Compared with some traditional coding algorithms such as the two-step weighted least squares algorithm, the Bayesian probability model is less affected by changes in background light intensity and can adapt well to the complex light environment in underground parking lots, thereby ensuring the accuracy and robustness of vehicle positioning.

[0075] This embodiment combines VLC positioning technology with the Bayesian probability model to achieve accurate positioning of vehicles in underground parking lots, effectively solving the problem that traditional positioning technology is difficult to use due to multiple environmental factors such as dim lighting, complex structure, and signal shielding in underground parking lots.

[0076] The vacant parking space identification module is used to detect the vacant parking space information in the underground parking lot in real time, and transmit the vacant parking space information in the underground parking lot to the display unit through the wireless transmission unit;

[0077] The empty parking space identification module includes a parking lot map unit, a parking lot space unit, a geomagnetic vehicle detector, a parking space status unit, a parking space information processing unit, and a parking space information transmission module; the parking lot map unit is implemented based on Tianbao TX8; the parking lot space unit is implemented based on EEPROM; the geomagnetic vehicle detector is implemented based on HES-DC01 of Huiershi; the parking space status unit is implemented based on STM32F103C8T6 single-chip microcomputer; the parking space information processing unit is implemented based on 51 single-chip microcomputer; and the parking space information transmission module is implemented based on HC-06 Bluetooth module.

[0078] The parking lot map unit stores map information of the underground parking lot where the vehicle is located;

[0079] The parking lot parking space unit stores the location information of all parking spaces in the underground parking lot;

[0080] The geomagnetic vehicle detector is installed on the ground of the parking space and is used to detect whether there is a vehicle parked in the parking space;

[0081] The vehicle status unit is used to receive the parking status of the vehicle detected by the geomagnetic vehicle detector, and mark the occupancy status of the parking space according to the parking status of the vehicle detected by the geomagnetic vehicle detector;

[0082] The parking space information processing unit determines whether there are vacant parking spaces in the underground parking lot, determines the positions of the vacant parking spaces, and calculates the number of vacant parking spaces based on the map information of the underground parking lot where the vehicle is located stored in the parking lot map unit, the position information of all parking spaces in the underground parking lot stored in the parking lot parking space unit, and the parking space occupancy status transmitted by the parking space status unit; the parking space information processing unit transmits the position information of the vacant parking spaces and the number information of the vacant parking spaces to the parking space information transmission module by wired transmission;

[0083] The parking space information transmission module is used to receive the position information of the empty parking spaces and the number information of the empty parking spaces from the parking space information processing unit, and transmit the information to the display unit through the wireless transmission unit.

[0084] The display unit is used to visualize the real-time position information of the vehicle transmitted by the VLC positioning device and the information of the available parking spaces in the underground parking lot transmitted by the empty parking space recognition module, and guide the driver to park the vehicle in the empty parking space;

[0085] The display unit includes an on-board display screen, which is used to visualize the real-time position information of the vehicle transmitted by the VLC positioning module and the position information and number information of the empty parking spaces transmitted by the parking space information transmission module, and guide the driver to park the vehicle in the empty parking space. The on-board display screen is implemented based on a TFT-LCD display screen.

[0086] The parking space number recognition device is used to recognize the parking space number information corresponding to the vehicle parked in the empty parking space, and receive the license plate number information transmitted by the front-end vehicle recognition device, and match the parking space number information with the license plate number information and transmit it to the mobile phone through the wireless transmission unit;

[0087] The parking space number recognition device includes a second camera, an image processing unit and a parking space number transmission unit; the second camera is implemented based on TC-CJ01; the image processing unit is implemented based on a development board of an STM32 single-chip microcomputer; and the parking space number transmission unit is implemented based on an HC-06 Bluetooth module.

[0088] The second camera is used to collect the rear environment information of the vehicle in the parking space, and transmit the collected rear environment information of the vehicle to the image processing unit; the second camera is installed on the ground at the rear of the parking space.

[0089] The image processing unit is used to receive the vehicle rear environment information transmitted by the second camera, and classify and extract the vehicle rear environment information through the KL dimension reduction kernel Fisher judgment parking space detection algorithm, so as to identify the parking space number information; firstly, the vehicle rear environment information is subjected to KL dimension reduction, so that the information in the vehicle rear environment information that is conducive to classification is concentrated in a small number of dimensions; secondly, the kernel Fisher discriminant analysis is used to perform parking space parking discrimination, and the parking space feature parameters are mapped to a new feature space through the kernel function, and then the Fisher linear discriminant analysis is performed, so as to identify the parking space number information and transmit it to the parking space number transmission unit;

[0090] The parking space number transmission unit is used to receive the parking space number information transmitted by the image processing unit and the license plate number information transmitted by the information transmission unit, and then match the parking space number information with the license plate number information to the driver's mobile phone through the wireless transmission unit;

[0091] The mobile phone is used to receive the license plate number information transmitted by the front-end vehicle identification device, the matching parking space number information and the license plate number information transmitted by the parking space number identification device, and assist the driver in finding an empty parking space in the underground parking lot, thereby finding the vehicle parked in the empty parking space; the mobile phone first receives the license plate number information transmitted by the front-end vehicle identification device, and after the vehicle parks in the empty parking space, the image processing unit of the parking space number identification device identifies the parking space number information, and matches it with the license plate number information transmitted by the front-end vehicle identification device, and transmits the matching license plate number information and parking space number information to the mobile phone again through the parking space number transmission unit, thereby ensuring that the license plate number information and the parking space number information correspond to each other, which helps the driver find the correct parking space.

[0092] The wireless transmission unit adopts wireless communication to assist the front-end vehicle identification device, VLC positioning device, empty parking space identification module, display unit, parking space number identification device, and information transmission between mobile phones. The present invention adopts NB-loT wireless communication mode, which has the characteristics of low power consumption and wide coverage.

[0093] See also Figure 2 , a method for locating vehicle information in an underground parking lot, specifically comprising the following steps:

[0094] S1. When a vehicle enters an underground parking lot, the front-end vehicle identification device detects the license plate number information, and transmits the license plate number information directly to the parking space number identification device and indirectly synchronizes it to the mobile phone through the wireless transmission unit;

[0095] S2. After the vehicle enters the underground parking lot, the VLC positioning device detects and locates the real-time position information of the vehicle, and transmits the real-time position information of the vehicle to the display unit through the wireless transmission unit;

[0096] The step S2 is specifically as follows:

[0097] Step S2-1: The VLC transmitter of the VLC positioning device emits visible light loaded with the lamp location information and ID information;

[0098] Step S2-2: The VLC receiver of the VLC positioning device receives the visible light loaded with the lamp location information and ID information in step S2-1, and first converts the optical signal into an electrical signal; then amplifies and filters the electrical signal, and uses a demodulation algorithm to convert the amplified and filtered electrical signal into a digital signal; finally, uses mean filtering to calculate the RSSI value corresponding to the received digital signal;

[0099] Step S2-3: After receiving the RSSI value in step S2-2, the VLC positioning module of the VLC positioning device establishes a conversion relationship model between RSSI and distance, and estimates the distance between the VLC receiver and each VLC transmitter; based on the estimated distances between more than three VLC receivers and each VLC transmitter and the known positions of the VLC transmitters, the real-time position information of the vehicle is calculated using a Bayesian probability model; the VLC positioning module transmits the real-time position information of the vehicle to the display unit through the wireless transmission unit.

[0100] S3, the vacant parking space recognition module detects the vacant parking space information in the underground parking lot, and transmits the vacant parking space information in the underground parking lot to the display unit through the wireless transmission unit;

[0101] The step S3 is specifically as follows:

[0102] Step S3-1: Acquire the map information of the underground parking lot stored in the parking lot map unit of the empty parking space recognition module;

[0103] Step S3-2: Acquire the location information of all parking spaces in the underground parking lot stored in the parking lot unit of the empty parking space identification module;

[0104] Step S3-3: obtaining information on vehicles parked in parking spaces detected by the geomagnetic vehicle detector of the empty parking space recognition module;

[0105] Step S3-4: the vehicle status unit of the empty parking space recognition module marks the occupancy status of the parking space according to the information of the vehicles parked in the parking space in step S3-3;

[0106] Step S3-5: the parking space information processing unit of the empty parking space identification module determines the location of the empty parking space and calculates the number of empty parking spaces according to the map information of the underground parking lot in step S3-1, the location information of all parking spaces in the underground parking lot in step S3-2, and the occupancy status of the parking spaces in step S3-4;

[0107] Step S3-6: After receiving the position information and number information of the empty parking spaces in step S3-5, the parking space information transmission module of the empty parking space identification module transmits the information to the display unit via the wireless transmission unit.

[0108] S4, the display unit visualizes the real-time position information of the vehicle transmitted by the VLC positioning device in step S2 and the real-time vacant parking space information transmitted by the vacant parking space recognition module in step S3 on the vehicle display screen, guiding the driver to park the vehicle in the designated vacant parking space;

[0109] S5. After the driver parks the vehicle in an empty parking space, the parking space number recognition device recognizes the parking space number information corresponding to the vehicle parked in the empty parking space, and combines the license plate number information transmitted by the front-end vehicle recognition device in step S1 to match the parking space number information and the license plate number information, and then transmits them to the mobile phone through the wireless transmission unit; the driver finds the vehicle parked in the empty parking space based on the matching parking space number information and license plate number information on the mobile phone.

[0110] The application effect of the present invention is described in detail below in conjunction with simulation experiments.

[0111] This simulation experiment considers different background light intensities and uses the control variable method to compare the effectiveness of the Bayesian probability model and the least squares method. The simulation is carried out in a digital underground parking lot model, which is designed according to the size and characteristics of the actual underground parking lot. In the simulation, the vehicle travels in a straight line at a constant speed to simulate the vehicle's driving trajectory under different light intensities.

[0112] Assume that the environmental parameters are in a parking lot with a size of 100mX100m. A VLC transmitter is installed every 5 meters on the top of the parking lot in a grid-like form. Each transmitter is 3.5 meters above the ground. The background light intensity is set to 300lux, 400lux, 500lux, and 600lux respectively. The multipath effect intensity is set to 0.1. The vehicle speed is 2km / h.

[0113] As shown in Table 1, the root mean square error (RMSE) values ​​of each positioning method under different background illumination intensities are listed in detail.

[0114] Table 1 Comparison of root mean square error (RMSE) values ​​between Bayesian probability model and least squares method under different background light intensities

[0115]

[0116] According to Table 1 above, when the vehicle is driving in an underground parking lot with a background light intensity of 300lux, 400lux, 500lux, and 600lux, respectively, the root mean square error (RMSE) value of the VLC positioning method based on the Bayesian probability model is smaller than the root mean square error (RMSE) value of the VLC positioning method based on the least squares method. The smaller the root mean square error (RMSE) value, the higher the prediction accuracy of the model, which indicates that the positioning accuracy of the vehicle is higher when the VLC positioning method under the Bayesian probability model is used.

[0117] Figure 3 (a)-(d) show the simulation results of estimating the vehicle's driving trajectory using the VLC positioning method under the least squares method and the VLC positioning method under the Bayesian probability model when the vehicle is traveling at a constant speed in an underground parking lot under the conditions of background light intensity of 300lux, 400lux, 500lux, and 600lux, respectively. It can be clearly seen that the degree of fit between the overall curve of the VLC positioning method under the Bayesian probability model and the actual vehicle driving trajectory is better than the degree of fit between the overall curve of the VLC positioning method under the least squares method and the actual vehicle driving trajectory; therefore, the deviation between the VLC positioning method under the least squares method and the actual position of the vehicle is large, while the deviation between the VLC positioning method under the Bayesian probability model and the actual position of the vehicle is small.

[0118] from Figure 3 From (c) and (d), we can find that with the increase of background light intensity, the RMSE values ​​of the VLC positioning method based on the least squares method and the VLC positioning method based on the Bayesian probability model change greatly, but the RMSE values ​​of the VLC positioning method based on the Bayesian probability model are all less than 0.5. Therefore, the VLC positioning method based on the Bayesian probability model can still maintain a relatively small and stable RMSE value under the condition of strong background light intensity, and has better accuracy and robustness.

[0119] observe Figure 4 From the root mean square error (RMSE) change curves of the two positioning algorithms, it can be found that: when the vehicle passes at a constant speed in an underground parking lot under the condition of a background light intensity of 500 lux, the value of the root mean square error change curve of the VLC positioning based on the least squares method is in the range of 0.5-0.7, and the value of the root mean square error change curve of the VLC positioning method based on the Bayesian probability model is in the range of 0.35-0.45; therefore, the value of the root mean square error change curve of the VLC positioning based on the Bayesian probability model is smaller, and the variation range is also small, which can maintain a higher positioning accuracy.

[0120] The above simulation verification shows that the VLC positioning method under the Bayesian probability model can maintain high positioning accuracy under different background light intensities and has good anti-interference and robustness.

[0121] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any modifications, equivalent substitutions and improvements made by any technician familiar with the technical field within the technical scope disclosed by the present invention and within the spirit and principle of the present invention should be covered by the protection scope of the present invention.

Claims

1. A vehicle positioning system for an underground parking lot, characterized by: It includes a front-end vehicle identification device, a VLC positioning device, an empty parking space identification module, a display unit, a parking space number identification device, a mobile phone, and a wireless transmission unit; The front-end vehicle identification device is used to detect the license plate number information of the vehicle entering the underground parking lot, and transmit the license plate number information directly to the parking space number identification device and indirectly synchronize it to the mobile phone through the wireless transmission unit; The VLC positioning device is used to detect and locate the position information of the vehicle entering the underground parking lot in real time, and transmit the real-time position information of the vehicle to the display unit through the wireless transmission unit; The vacant parking space identification module is used to detect the vacant parking space information in the underground parking lot in real time, and transmit the vacant parking space information in the underground parking lot to the display unit through the wireless transmission unit; The display unit is used to visualize the real-time position information of the vehicle transmitted by the VLC positioning device and the information of the available parking spaces in the parking lot transmitted by the empty parking space recognition module, and guide the driver to park the vehicle in the empty parking space; The parking space number recognition device is used to recognize the parking space number information corresponding to the vehicle parked in the empty parking space, and receive the license plate number information transmitted by the front-end vehicle recognition device, and match the parking space number information with the license plate number information and transmit it to the mobile phone through the wireless transmission unit; The mobile phone is used to receive the license plate number information transmitted by the front-end vehicle identification device, the matching parking space number information and the license plate number information transmitted by the parking space number identification device, and assist the driver in finding an empty parking space in the underground parking lot, thereby finding the vehicle parked in the empty parking space; The wireless transmission unit adopts wireless communication to assist the front-end vehicle identification device, the VLC positioning device, the empty parking space identification module, the display unit, the parking space number identification device, and the information transmission between the mobile phone.

2. The underground parking lot vehicle information positioning system according to claim 1, characterized in that: The front-end vehicle identification device includes a first camera, an information processing unit, and an information transmission unit; The first camera is used to collect images of vehicles entering the underground parking lot and transmit the collected vehicle images to the information processing unit; The information processing unit recognizes and processes the vehicle image captured by the first camera, thereby obtaining the vehicle license plate number information; After receiving the license plate number information of the vehicle acquired by the information processing unit, the information transmission unit directly transmits the license plate number information to the parking space number transmission unit of the parking space number recognition device and indirectly synchronizes it to the mobile phone through the wireless transmission unit.

3. The underground parking lot vehicle information positioning system according to claim 1, characterized in that: The VLC positioning device comprises a VLC transmitter, a VLC receiver, and a VLC positioning module; The VLC transmitter includes a plurality of LED lamps and a digital signal processor; the digital signal processor sequentially pre-processes, encodes, and modulates the pulse signal, and uses the modulated pulse signal to control the LED lamp to emit visible light loaded with the lamp position information and ID information; The VLC receiver includes a photoelectric converter and a signal processing module; after receiving the visible light loaded with the lamp position information and ID information, the photoelectric converter converts the optical signal into an electrical signal; The signal processing module amplifies and filters the received electrical signal, and then uses a demodulation algorithm to convert the amplified and filtered electrical signal into a digital signal, and uses mean filtering to calculate the RSSI value corresponding to the received digital signal; finally, the RSSI information is transmitted to the VLC positioning module through wired transmission; The VLC positioning module is used to receive the RSSI value of the VLC receiver, establish a conversion relationship model between RSSI and distance, and estimate the distance between the VLC receiver and each VLC transmitter; based on the estimated distances between more than three VLC receivers and each VLC transmitter and the known positions of the VLC transmitters, the real-time position information of the vehicle is calculated using a Bayesian probability model; the VLC positioning module transmits the real-time position information of the vehicle to a display unit through a wireless transmission unit.

4. The underground parking lot vehicle information positioning system according to claim 3, characterized in that: The method of using the Bayesian probability model to calculate the real-time position information of the vehicle specifically includes: assuming that the initial position of the vehicle obeys a Gaussian distribution, predicting the position of the vehicle at the next moment according to the vehicle motion model and the historical trajectory at each time step; updating the prior probability according to the predicted position of the vehicle at the next moment; calculating the likelihood probability according to the RSSI value received by the VLC receiver at each time step; updating the posterior probability distribution based on the calculated prior probability and likelihood probability, normalizing the posterior probability distribution, and estimating the real-time position of the vehicle through the normalized posterior probability distribution.

5. The underground parking lot vehicle information positioning system according to claim 1, characterized in that: The empty parking space identification module includes a parking lot map unit, a parking lot space unit, a geomagnetic vehicle detector, a parking space status unit, a parking space information processing unit, and a parking space information transmission module; The parking lot map unit stores map information of the underground parking lot where the vehicle is located; The parking lot parking space unit stores the location information of all parking spaces in the underground parking lot; The geomagnetic vehicle detector is used to detect whether there is a vehicle parked in the parking space; The vehicle status unit is used to receive the parking status of the vehicle detected by the geomagnetic vehicle detector, and mark the occupancy status of the parking space according to the parking status of the vehicle detected by the geomagnetic vehicle detector; The parking space information processing unit determines whether there are vacant parking spaces in the parking lot, determines the locations of the vacant parking spaces, and calculates the number of vacant parking spaces based on the map information of the underground parking lot where the vehicle is located stored in the parking lot map unit, the location information of all parking spaces in the underground parking lot stored in the parking lot parking space unit, and the parking space occupancy status transmitted by the parking space status unit; The parking space information processing unit transmits the location information of the empty parking spaces and the number information of the empty parking spaces to the parking space information transmission module by wired transmission; The parking space information transmission module is used to receive the position information of the empty parking spaces and the number information of the empty parking spaces from the parking space information processing unit, and transmit the information to the display unit through the wireless transmission unit.

6. An underground parking lot vehicle information positioning system according to claim 3 or 5, characterized in that: The display unit includes an on-board display screen, which is used to visualize the real-time position information of the vehicle transmitted by the VLC positioning module and the position information and number information of the empty parking spaces transmitted by the parking space information transmission module, so as to guide the driver to park the vehicle in the empty parking space.

7. An underground parking lot vehicle information positioning system according to claim 1 or 2, characterized in that: The parking space number recognition device includes a second camera, an image processing unit and a parking space number transmission unit; The second camera is used to collect the rear environment information of the vehicle in the parking space, and transmit the collected rear environment information of the vehicle to the image processing unit; The image processing unit is used to receive the vehicle rear environment information transmitted by the second camera, and classify and extract the vehicle rear environment information through the KL dimension reduction kernel Fisher judgment parking space detection algorithm, so as to identify the parking space number information; firstly, the vehicle rear environment information is subjected to KL dimension reduction, so that the information in the vehicle rear environment information that is conducive to classification is concentrated in a small number of dimensions; secondly, the kernel Fisher discriminant analysis is used to perform parking space parking discrimination, and the parking space feature parameters are mapped to a new feature space through the kernel function, and then the Fisher linear discriminant analysis is performed, so as to identify the parking space number information and transmit it to the parking space number transmission unit; The parking space number transmission unit is used to receive the parking space number information transmitted by the image processing unit and the license plate number information transmitted by the information transmission unit, and then match the parking space number information with the license plate number information and transmit them to the driver's mobile phone through the wireless transmission unit.

8. A method for locating vehicle information in an underground parking lot, characterized by: The specific steps include: S1. When a vehicle enters an underground parking lot, the front-end vehicle identification device detects the license plate number information, and transmits the license plate number information directly to the parking space number identification device and indirectly synchronizes it to the mobile phone through the wireless transmission unit; S2. After the vehicle enters the underground parking lot, the VLC positioning device detects and locates the real-time position information of the vehicle, and transmits the real-time position information of the vehicle to the display unit through the wireless transmission unit; S3, the vacant parking space recognition module detects the vacant parking space information in the underground parking lot, and transmits the vacant parking space information in the underground parking lot to the display unit through the wireless transmission unit; S4, the display unit visualizes the real-time position information of the vehicle transmitted by the VLC positioning device in step S2 and the real-time vacant parking space information transmitted by the vacant parking space recognition module in step S3 on the vehicle display screen, guiding the driver to park the vehicle in the designated vacant parking space; S5. After the driver parks the vehicle in an empty parking space, the parking space number recognition device recognizes the parking space number information corresponding to the vehicle parked in the empty parking space, and combines the license plate number information transmitted by the front-end vehicle recognition device in step S1 to match the parking space number information and the license plate number information, and then transmits them to the mobile phone through the wireless transmission unit; the driver searches for an empty parking space based on the matching parking space number information and license plate number information on the mobile phone, thereby finding the vehicle parked in the empty parking space.

9. The method for locating vehicle information in an underground parking lot according to claim 8, characterized in that: The step S2 is specifically as follows: Step S2-1: The VLC transmitter of the VLC positioning device emits visible light loaded with the lamp location information and ID information; Step S2-2: The VLC receiver of the VLC positioning device receives the visible light loaded with the lamp location information and ID information in step S2-1, and first converts the optical signal into an electrical signal; then amplifies and filters the electrical signal, and uses a demodulation algorithm to convert the amplified and filtered electrical signal into a digital signal; finally, uses mean filtering to calculate the RSSI value corresponding to the received digital signal; Step S2-3: After receiving the RSSI value in step S2-2, the VLC positioning module of the VLC positioning device establishes a conversion relationship model between RSSI and distance, and estimates the distance between the VLC receiver and each VLC transmitter; based on the estimated distances between more than three VLC receivers and each VLC transmitter and the known positions of the VLC transmitters, the real-time position information of the vehicle is calculated using a Bayesian probability model; the VLC positioning module transmits the real-time position information of the vehicle to the display unit through the wireless transmission unit.

10. The method for locating vehicle information in an underground parking lot according to claim 8, characterized in that: The step S3 is specifically as follows: Step S3-1: Acquire the map information of the underground parking lot stored in the parking lot map unit of the empty parking space recognition module; Step S3-2: Acquire the location information of all parking spaces in the underground parking lot stored in the parking lot unit of the empty parking space identification module; Step S3-3: obtaining information on vehicles parked in parking spaces detected by the geomagnetic vehicle detector of the empty parking space recognition module; Step S3-4: the vehicle status unit of the empty parking space recognition module marks the occupancy status of the parking space according to the information of the vehicles parked in the parking space in step S3-3; Step S3-5: the parking space information processing unit of the empty parking space identification module determines the location of the empty parking space and calculates the number of empty parking spaces according to the map information of the underground parking lot in step S3-1, the location information of all parking spaces in the underground parking lot in step S3-2, and the occupancy status of the parking spaces in step S3-4; Step S3-6: After receiving the position information and the number information of the empty parking spaces in step S3-5, the parking space information transmission module of the empty parking space identification module transmits the information to the display unit via the wireless transmission unit.

Images