CN114550425A - Pedestrian perception vehicle-road cooperative early warning device and method based on millimeter wave radar - Google Patents
Pedestrian perception vehicle-road cooperative early warning device and method based on millimeter wave radar Download PDFInfo
- Publication number
- CN114550425A CN114550425A CN202111278474.9A CN202111278474A CN114550425A CN 114550425 A CN114550425 A CN 114550425A CN 202111278474 A CN202111278474 A CN 202111278474A CN 114550425 A CN114550425 A CN 114550425A
- Authority
- CN
- China
- Prior art keywords
- data
- pedestrian
- millimeter wave
- wave radar
- early warning
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000008447 perception Effects 0.000 title claims abstract description 20
- 238000000034 method Methods 0.000 title claims abstract description 15
- 238000004364 calculation method Methods 0.000 claims abstract description 34
- 240000007651 Rubus glaucus Species 0.000 claims abstract description 15
- 235000011034 Rubus glaucus Nutrition 0.000 claims abstract description 15
- 235000009122 Rubus idaeus Nutrition 0.000 claims abstract description 15
- 230000006870 function Effects 0.000 claims abstract description 10
- 238000011161 development Methods 0.000 claims abstract description 9
- 238000004891 communication Methods 0.000 claims abstract description 4
- 238000013136 deep learning model Methods 0.000 claims description 10
- 238000012216 screening Methods 0.000 claims description 5
- 230000004913 activation Effects 0.000 claims description 3
- 230000004927 fusion Effects 0.000 claims 1
- 230000008901 benefit Effects 0.000 abstract description 6
- 238000001514 detection method Methods 0.000 abstract description 3
- 238000013527 convolutional neural network Methods 0.000 description 12
- 230000018109 developmental process Effects 0.000 description 5
- 238000012549 training Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000013526 transfer learning Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/18—Status alarms
- G08B21/24—Reminder alarms, e.g. anti-loss alarms
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/93—Radar or analogous systems specially adapted for specific applications for anti-collision purposes
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0108—Measuring and analyzing of parameters relative to traffic conditions based on the source of data
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Electromagnetism (AREA)
- Computer Networks & Wireless Communication (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention provides a pedestrian perception vehicle-road cooperative early warning device and method based on a millimeter wave radar, wherein a radar data acquisition module acquires pedestrian data through the millimeter wave radar and transmits the pedestrian data to an edge calculation module in real time through a connection port; the edge calculation module carries out real-time identification operation on pedestrian data through a Jetson Nano development board to obtain the pedestrian distance and the ID; the data unloading module receives the identification result sent from the Jetson Nano in a WIFI wireless communication mode through the raspberry group, uploads the identification result to the MQTT server, and stores the identification result locally; the MQTT server receives data transmitted from the raspberry group and sends the data to the data display module; the data display module displays pedestrian distance and ID data acquired from the MQTT server through the vehicle-mounted terminal APP and judges whether operation is needed. The invention has the advantages of small volume, convenient connection and debugging, simple structure, accurate identification result, low cost and the like, can effectively carry out real-time target detection and achieves better cooperative vehicle and road early warning function.
Description
Technical Field
The invention relates to the field of vehicle-road cooperative early warning, in particular to a pedestrian perception vehicle-road cooperative early warning device and method based on a millimeter wave radar.
Background
The safety of pedestrian crossing the street is always an important problem in the field of traffic safety, and sometimes even if drivers concentrate their efforts, it is difficult to predict pedestrians suddenly appearing in the blind area of the intersection. At present, the early warning means for reminding a driver in real time is limited, and the driver cannot sense and warn in advance when crossing a street. With the development of urbanization, the country gradually pays attention to the iteration of the vehicle-road cooperative new technology, but at present, no mature vehicle-road cooperative pedestrian early warning device is applied. The main means for sensing the pedestrians in the existing pedestrian sensing vehicle-road cooperative early warning is a camera, but the camera is a visible light sensor, is a sensor based on a visible light source, is easily interfered by the external environment, and has the defects of large data volume, complex processing, short recognition distance, need to occupy a large amount of processor resources and the like. In contrast, the millimeter wave radar senses more parameter types and has higher data processing speed. Therefore, the invention provides a pedestrian perception vehicle-road cooperative early warning device based on a millimeter wave radar.
Disclosure of Invention
The invention aims to provide a pedestrian perception vehicle-road cooperative early warning device and method based on a millimeter wave radar.
The technical solution for realizing the purpose of the invention is as follows: the utility model provides a pedestrian perception vehicle and road cooperative early warning device based on millimeter wave radar, includes radar data acquisition module, edge calculation module, data unloading module, MQTT server and data display module, wherein: the radar data acquisition module acquires pedestrian data through a millimeter wave radar and transmits the pedestrian data to the edge calculation module in real time through a connection port; the edge calculation module carries out real-time identification operation on pedestrian data through a Jetson Nano development board to obtain the pedestrian distance and the ID; the data unloading module receives the identification result sent from the Jetson Nano in a WIFI wireless communication mode through the raspberry group, uploads the identification result to the MQTT server, and stores the identification result locally; the MQTT server receives data transmitted from the raspberry group and sends the data to the data display module; the data display module displays pedestrian distance and ID data acquired from the MQTT server through the vehicle-mounted terminal APP and judges whether operation is needed.
Furthermore, the radar data acquisition module comprises a millimeter wave radar and a CAN bus analyzer, wherein the millimeter wave radar is connected with the bus analyzer through a CAN bus, the CAN bus analyzer is connected with a CAN bus to be connected with a Jetsonn Nano suite through a USB interface, the millimeter wave radar acquires pedestrian state data, and the data are transmitted to the edge calculation module through a USB interface in real time.
Further, the edge calculation module is developed based on a Jetson Nano suite, 6 kinds of data in radar original data are screened out, the data comprise standard deviation Orientation _ rms of a deviation angle of a target, a longitudinal speed VrelLong of the target, a transverse distance DistLat of the target, a length of the target, a reflection energy value RCS of the target and an experience classification Class fused with the radar, and the six kinds of data are deployed on a CNN deep learning model trained on a Jetson Nano development board to perform real-time identification operation.
Furthermore, the edge calculation module fills in missing values by using forward interpolation before screening data.
Furthermore, the CNN deep learning model in the edge calculation module has 5 layers in total, and the first layer is an input layer and includes 64 convolution kernels with a length of 8; the second layer and the third layer are convolution layers, each layer comprises 128 convolution kernels with the length of 5, and the activation functions of the convolution kernels are ReLU functions; the fourth layer is a batch normalization layer, and data is standardized; the fifth layer is a softmax layer and outputs the classification result.
Furthermore, before performing CNN deep learning model identification operation, the edge calculation module expands the data after real-time screening through a sliding window, where the window width is 6 and the sliding distance is 5.
Further, the data unloading module is developed based on raspberry group 4B, obtains the identification result from the edge calculation module, and uploads the identification result to the MQTT server, wherein the uploading tool uses a 4G public network and adopts a 909s-821LTE module to prepare for sending to the vehicle-mounted terminal in the next step.
Further, the vehicle-mounted terminal APP is mounted on an android mobile phone or a tablet.
Furthermore, the radar data acquisition module is arranged on a pedestrian traffic light pole at the intersection and faces the pedestrian crosswalk, and the edge calculation module is connected with the radar data acquisition module in a wired mode and arranged above the traffic light pole; the data unloading module and the edge calculation module transmit information in a WiFi mode and are arranged above a traffic signal lamp post; the data display module is placed in the vehicle, and a driver can conveniently check early warning information in real time.
A pedestrian perception vehicle-road cooperative early warning method based on a millimeter wave radar is based on the pedestrian perception vehicle-road cooperative early warning system based on the millimeter wave radar, and pedestrian perception vehicle-road cooperative early warning based on the millimeter wave radar is achieved.
Compared with the prior art, the invention has the following remarkable advantages: 1) the Jetsonnano is used as the identified hardware resource, so that the transmitted original radar data can be directly and rapidly identified in real time without sending the data to a third-party server for identification. 2) The device can perform early warning in real time through sensing and communication, and a user can directly use a mobile phone APP to receive the information of the MQTT server side. 3) Aiming at the classification problem of the millimeter wave radar, 6 features are selected as the input of a classifier, the data classification problem is converted into a time series classification problem, and a sliding window method is introduced as a data enhancement technology, so that the classification method has a good classification effect on pedestrians.
Drawings
FIG. 1 is an overall architecture diagram of the pedestrian-aware vehicle-road cooperative early warning device based on millimeter wave radar according to the present invention;
fig. 2 is a working principle diagram of the pedestrian perception vehicle-road cooperative early warning device based on the millimeter wave radar.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
The invention provides a pedestrian perception vehicle-road cooperative early warning device based on a millimeter wave radar and an embedded artificial intelligence chip, as shown in figure 1, comprising: the device comprises a radar data acquisition module, an edge calculation module, a data unloading module, an MQTT server and a data display module. The radar data acquisition module is used for acquiring an original data set of a radar; the edge calculation module is used for identifying pedestrian data in real time; the data unloading module is used for forwarding and storing the identification result; the MQTT server is used for uploading and issuing identification results; the data display module is used for sending the early warning message to the vehicle-mounted terminal APP, so that the user can obtain the early warning result of the pedestrian information and process the early warning result in time.
Among the above-mentioned modules, radar data acquisition module can install in the pedestrian traffic signal lamp pole at intersection, towards the pedestrian crossing. The edge calculation module is developed based on a Jetson Nano suite, is connected with the radar data acquisition module in a wired mode, and is also installed above a lamp post of a traffic signal lamp. The data unloading module is based on raspberry group 4B development, and with marginal calculation module through wiFi mode transmission information, install in traffic signal lamp pole top. The data display module is placed in the vehicle, and a driver can conveniently check early warning information in real time.
The following is an introduction of the optimized design of each module.
(I) radar data acquisition module
The radar data acquisition module comprises a millimeter wave radar and a CAN bus analyzer, wherein the millimeter wave radar is connected with the bus analyzer through a CAN bus, the CAN bus analyzer is connected with a CAN bus to be connected with a Jetson Nano suite through a USB interface, the pedestrian state data are acquired by the millimeter wave radar, and the data are transmitted to the edge calculation module through the USB interface in real time. In the invention, the millimeter wave radar can use a 77GHz long-medium distance millimeter wave radar ARS408 in Germany continental. The acquired data includes 1) RCS: a reflected energy value of the target; 2) DistLat: the lateral distance of the target; 3) DistLong: the longitudinal distance of the target; 4) VrleLong: a target longitudinal velocity;
5) VeleLat: a target lateral velocity; 6) length: the length of the target; 7) width: the width of the target, etc. for a total of 27 radar raw data.
(II) edge calculation module
And the edge calculation module performs real-time identification operation on the trained CNN deep learning model on the Jetson Nano suite, and finally obtains an identification result comprising the pedestrian distance and the ID. For example, the recognition result indicates that "a pedestrian is identified with ID 001, a pedestrian is identified with distance 10.8 m", "a pedestrian is identified with ID 002, a pedestrian is identified with distance 11 m", "a pedestrian is identified with ID 003, and a pedestrian is identified with distance 13 m". In the invention, a Jetson Nano suite is adopted by the edge calculation module. The Jetson Nano suite comprises a 4-core A57CPU, a 128-core Maxwell architecture GPU and a 4G memory. The Jetson Nano has the greatest advantage of volume, the core board is detachably designed, the size of the core board is only 70x 45mm, the core board can be conveniently integrated in various embedded applications, and meanwhile, the power consumption of the core board is very low. The algorithm flow of recognition is briefly described below:
first, preprocessing the received data on Jetson nano: 1) processing missing values: because about 15 pieces of object information can be collected in the millimeter wave radar 1s, and only a few parts change at different time, the adopted method is forward interpolation to fill in missing values. 2) 6 data of the 27 data are screened out, including standard deviation of deviation angle of the target, Orientation _ rms, longitudinal speed of the target, VrelLong, transverse distance of the target, DistLat, length of the target, reflection energy value RCS of the target and empirical classification of the radar itself. The six kinds of data are deployed on a CNN deep learning model trained on a Jetson Nano suite to carry out real-time recognition operation.
And secondly, in order to identify the target, a deep learning model needs to be built, and the method uses a customized CNN model with 5 layers in total. The first layer is the input layer, which uses as input the time series of screened seed radar data, which contains 64 convolution kernels of length 8. The second and third layers are convolutional layers, each containing 128 convolutional kernels of length 5, whose activation functions are both ReLU functions. The fourth layer is the BatchNormalization layer, which normalizes the data. The fifth layer is a softmax layer and outputs the classification result.
And thirdly, data enhancement, wherein data obtained by the radar in real-time measurement is much less than that obtained in training, so that a sliding window needs to be established, the data subjected to real-time screening is expanded through the sliding window to enable the CNN network to identify, and because 6 characteristics of the detected object have strong correlation with time, the relation between the data can be enhanced through the sliding window to obtain a more accurate identification effect. A large amount of data of automobiles, pedestrians and non-motor vehicles are measured in advance by using an ARS408 millimeter wave radar, the measured data are divided into a training set and a verification set, the ratio is 8.5:1.5, and the training set is sent to a CNN network for training. After the parameters of the sliding window are adjusted for many times, the final parameters are determined to be the length of the window of 15, and because only 6 kinds of data are selected, the width of the window is 6, the sliding distance is 5, and the real-time performance and the precision of the neural network established by the method are the best under the parameters.
(III) data unloading module
And the data unloading module is developed based on the raspberry pi 4B, acquires the identification result from the edge calculation module, uploads the identification result to the MQTT server, and prepares for sending to the user mobile terminal in the next step. The uploading tool can use a 4G public network, adopts a Hua Shi 909s-821LTE module and is connected with the raspberry pi through a USB interface. In the invention, the data unloading module adopts a raspberry pi 4B, and the raspberry pi 4B adopts a quad-core ARM Cortex-A72, 1.5GHz processor; the USB interface card comprises 4 USB interfaces, an extended 40-pin GPIO plug and an onboard BCM43143WiFi chip, and supports Bluetooth and WIFI; an HDMI output port; a storage device: a microSD card; an SD card reader.
(IV) data display module
And the data display module sends the uploaded identification result to an APP (application) of the vehicle-mounted terminal, so that a user can timely acquire the identification result acquired from the MQTT server side and timely judge whether to carry out operations such as avoidance in advance. The data display module can be installed on an APP of a mobile phone or a tablet at a vehicle-mounted end.
In conclusion, the pedestrian perception vehicle-road cooperative early warning device is realized based on the millimeter wave radar and the embedded artificial intelligence chip, and provides radar original data acquisition; identifying the pedestrian; storing and forwarding data to a server; and functions of user data display, early warning information acquisition and the like are achieved.
Based on the above, the invention also provides a pedestrian perception vehicle-road cooperative early warning method, which comprises the steps that firstly, a radar data acquisition module acquires radar original data; the edge calculation module obtains radar original data from the radar data module data, performs identification operation through a CNN deep learning model, and then sends an identification result to the data dump module; the data unloading module transmits the identification result data to an MQTT server through an MQTT protocol; and finally, the data display module sends the user mobile phone APP, the user acquires pedestrian distance information and the corresponding ID, the pedestrian distance information and the corresponding ID can be processed in time, and a specific implementation scheme can be designed as follows:
radar raw data acquisition: after the millimeter wave radar is started to operate, analyzing the CAN message by adopting Python language to obtain the original data of the millimeter wave radar.
And (3) pedestrian identification: a transfer learning model of a TSC (time series identification model) problem based on CNN is adopted, and the main structure of the network is a single-dimensional convolution neural architecture. The input of the network device is a group of multi-feature time sequences, the output of the network device is a pedestrian target, the time sequence tracked by the millimeter wave radar is subjected to sliding window segmentation, and the network is trained on the subsequence, so that the data enhancement effect is obvious on the one hand. On the other hand, the classification of the corresponding target is identified in the shortest time through the division of the short time sequence, so that the real-time performance of the algorithm is enhanced
And (3) data storage: the JetsonnNano suite receives the analyzed original data, continuously adds the analyzed original data into a newly-built txt file through Python language, and stores the txt file in the SD card, so that a transfer learning model of the TSC (time series identification model) problem based on CNN (CNN) embedded in the SD card can read the data in the txt file and carry out identification and classification to generate packed data (pedestrian ID and distance).
Data forwarding to the server: after the packaged data sent from the edge computing module is received by the WiFi module, the packaged data can be stored locally (raspberry pi 4B). The raspberry group uploads data to an MQTT server deployed at the cloud end through an LTE module via a 4G network, and the server is responsible for transferring the data.
The method comprises the following steps of user data display and early warning information acquisition: the user obtains the pedestrian ID and the distance information issued by the MQTT server through the mobile phone APP, and the real-time alarm function can be achieved.
In summary, the present invention has the following advantages: the device has the advantages of small volume, convenient connection and debugging, simple structure, accurate recognition result and low cost, is suitable for development and utilization of real-time pedestrian detection and early warning, can effectively carry out real-time target detection, and achieves better vehicle-road cooperative early warning function.
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (10)
1. The utility model provides a pedestrian perception vehicle and road cooperative early warning device based on millimeter wave radar, its characterized in that includes radar data acquisition module, marginal calculation module, data unloading module, MQTT server and data display module, wherein: the radar data acquisition module acquires pedestrian data through a millimeter wave radar and transmits the pedestrian data to the edge calculation module in real time through a connection port; the edge calculation module carries out real-time identification operation on pedestrian data through a Jetson Nano development board to obtain the pedestrian distance and the ID; the data unloading module receives the identification result sent from the Jetson Nano in a WIFI wireless communication mode through the raspberry group, uploads the identification result to the MQTT server, and stores the identification result locally; the MQTT server receives data transmitted from the raspberry group and sends the data to the data display module; the data display module displays pedestrian distance and ID data acquired from the MQTT server through the vehicle-mounted terminal APP and judges whether operation is needed.
2. The pedestrian-aware vehicle-road cooperative early warning device based on the millimeter wave radar as claimed in claim 1, wherein the radar data acquisition module comprises the millimeter wave radar and a CAN bus analyzer, wherein the millimeter wave radar is connected with the bus analyzer through a CAN bus, the CAN bus analyzer is connected with a Jetson Nano suite through a CAN bus-USB interface, the millimeter wave radar acquires pedestrian state data, and transmits the data to the edge calculation module through a USB interface in real time.
3. The pedestrian perception vehicle-road cooperative early warning device based on the millimeter wave radar as claimed in claim 1, wherein the edge calculation module screens 6 kinds of data in radar raw data based on Jetson Nano suite development, wherein the 6 kinds of data include a standard deviation order _ rms of a deviation angle of a target, a longitudinal speed VrelLong of the target, a transverse distance DistLat of the target, a length h of the target, a reflection energy value RCS of the target and an experience classification Class of fusion of the radar, and the six kinds of data are deployed on a CNN deep learning model trained on a Jetson Nano development board to perform real-time recognition operation.
4. The pedestrian-aware vehicle-road cooperative early warning device based on the millimeter wave radar as claimed in claim 3, wherein the edge calculation module fills the missing value by adopting forward interpolation before screening data.
5. The pedestrian perception vehicle-road cooperative early warning device based on the millimeter wave radar as claimed in claim 3, wherein the CNN deep learning model in the edge calculation module has 5 layers in total, and the first layer is an input layer and comprises 64 convolution kernels with the length of 8; the second layer and the third layer are convolution layers, each layer comprises 128 convolution kernels with the length of 5, and the activation functions of the convolution kernels are ReLU functions; the fourth layer is a BatchNormalization layer, and data are standardized; the fifth layer is a softmax layer and outputs the classification result.
6. The pedestrian-aware vehicle-road cooperative early warning device based on the millimeter wave radar as claimed in claim 3, wherein the edge calculation module expands the data after real-time screening through a sliding window before performing CNN deep learning model identification operation, the window width is 6, and the sliding distance is 5.
7. The pedestrian perception vehicle-road cooperative early warning device based on the millimeter wave radar as claimed in claim 1, wherein the data unloading module is developed based on raspberry pi 4B, obtains the recognition result from the edge calculation module, and uploads the recognition result to the MQTT server, and the uploading tool uses a 4G public network and adopts a 909s-821LTE module to prepare for sending to the vehicle-mounted terminal in the next step.
8. The pedestrian perception vehicle-road cooperative early warning device based on the millimeter wave radar of claim 1, wherein the vehicle-mounted terminal APP is mounted on an android mobile phone or a tablet.
9. The pedestrian perception vehicle-road cooperative early warning device based on the millimeter wave radar as claimed in claim 1, wherein the radar data acquisition module is mounted on a pedestrian traffic signal lamp pole at an intersection, faces a pedestrian crosswalk, and the edge calculation module is connected with the radar data acquisition module in a wired manner and mounted above the traffic signal lamp pole; the data unloading module and the edge calculation module transmit information in a WiFi mode and are arranged above a traffic signal lamp post; the data display module is placed in the vehicle, and a driver can conveniently check early warning information in real time.
10. A pedestrian-aware vehicle-road cooperative early warning method based on millimeter wave radar is characterized in that pedestrian-aware vehicle-road cooperative early warning based on millimeter wave radar is realized based on the pedestrian-aware vehicle-road cooperative early warning system based on millimeter wave radar of any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111278474.9A CN114550425A (en) | 2021-10-30 | 2021-10-30 | Pedestrian perception vehicle-road cooperative early warning device and method based on millimeter wave radar |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111278474.9A CN114550425A (en) | 2021-10-30 | 2021-10-30 | Pedestrian perception vehicle-road cooperative early warning device and method based on millimeter wave radar |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114550425A true CN114550425A (en) | 2022-05-27 |
Family
ID=81668806
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111278474.9A Pending CN114550425A (en) | 2021-10-30 | 2021-10-30 | Pedestrian perception vehicle-road cooperative early warning device and method based on millimeter wave radar |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114550425A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190204834A1 (en) * | 2018-01-04 | 2019-07-04 | Metawave Corporation | Method and apparatus for object detection using convolutional neural network systems |
| CN112328651A (en) * | 2020-10-19 | 2021-02-05 | 南京理工大学 | Traffic target identification method based on millimeter wave radar data statistical characteristics |
| CN112906019A (en) * | 2021-02-04 | 2021-06-04 | 南京邮电大学 | Flow data generation method, device and system based on improved DCGAN model |
| CN112926773A (en) * | 2021-02-23 | 2021-06-08 | 深圳市北斗智能科技有限公司 | Riding safety early warning method and device, electronic equipment and storage medium |
| CN113479218A (en) * | 2021-08-09 | 2021-10-08 | 哈尔滨工业大学 | Roadbed automatic driving auxiliary detection system and control method thereof |
-
2021
- 2021-10-30 CN CN202111278474.9A patent/CN114550425A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190204834A1 (en) * | 2018-01-04 | 2019-07-04 | Metawave Corporation | Method and apparatus for object detection using convolutional neural network systems |
| CN112328651A (en) * | 2020-10-19 | 2021-02-05 | 南京理工大学 | Traffic target identification method based on millimeter wave radar data statistical characteristics |
| CN112906019A (en) * | 2021-02-04 | 2021-06-04 | 南京邮电大学 | Flow data generation method, device and system based on improved DCGAN model |
| CN112926773A (en) * | 2021-02-23 | 2021-06-08 | 深圳市北斗智能科技有限公司 | Riding safety early warning method and device, electronic equipment and storage medium |
| CN113479218A (en) * | 2021-08-09 | 2021-10-08 | 哈尔滨工业大学 | Roadbed automatic driving auxiliary detection system and control method thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112417967B (en) | Obstacle detection method, obstacle detection device, computer device, and storage medium | |
| US11043005B2 (en) | Lidar-based multi-person pose estimation | |
| WO2021134172A1 (en) | Trajectory prediction method and related device | |
| US11327178B2 (en) | Piece-wise network structure for long range environment perception | |
| CN104853972A (en) | Augmenting ADAS features of vehicle with image processing support in on-board vehicle platform | |
| CN113155173B (en) | Perception performance evaluation method and device, electronic device and storage medium | |
| CN113962141A (en) | Target detection model automatic iteration method, device and storage medium | |
| CN112580571A (en) | Vehicle running control method and device and electronic equipment | |
| CN104168308A (en) | Internet-of-vehicles inspecting system based on cloud computation and mass data | |
| CN116347385A (en) | Communication perception and edge calculation integrated system for vehicle-road cooperation | |
| CN110083099B (en) | Automatic driving architecture system meeting automobile function safety standard and working method | |
| CN203933691U (en) | Based on the vehicle auditing system of car networking | |
| CN110909656B (en) | Pedestrian detection method and system integrating radar and camera | |
| CN117372979A (en) | Road inspection method, device, electronic equipment and storage medium | |
| CN109808600A (en) | The method for visualizing and device of the perception information of automatic driving vehicle | |
| Kheder et al. | Real-time traffic monitoring system using IoT-aided robotics and deep learning techniques | |
| CN114550425A (en) | Pedestrian perception vehicle-road cooperative early warning device and method based on millimeter wave radar | |
| US11747454B2 (en) | Granularity-flexible existence-based object detection | |
| CN115421122A (en) | Target object detection method and device, electronic equipment and readable storage medium | |
| CN110556024B (en) | Anti-collision auxiliary driving method and system and computer readable storage medium | |
| Cao et al. | Development of Intelligent Multimodal Traffic Monitoring using Radar Sensor at Intersections | |
| CN206856690U (en) | A kind of intersection collision avoidance system based on RFID system | |
| CN111354205A (en) | ETC-based road condition information acquisition method and system, storage medium and intelligent terminal | |
| CN111542828A (en) | Line recognition method, line recognition device, line recognition system, and computer storage medium | |
| CN116434041B (en) | Mining method, device and equipment for error perception data and automatic driving vehicle |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |