WO2023108932A1 - 一种基于毫米波雷达的车辆异常行驶行为识别方法 - Google Patents

一种基于毫米波雷达的车辆异常行驶行为识别方法 Download PDF

Info

Publication number
WO2023108932A1
WO2023108932A1 PCT/CN2022/081190 CN2022081190W WO2023108932A1 WO 2023108932 A1 WO2023108932 A1 WO 2023108932A1 CN 2022081190 W CN2022081190 W CN 2022081190W WO 2023108932 A1 WO2023108932 A1 WO 2023108932A1
Authority
WO
WIPO (PCT)
Prior art keywords
vehicle
compare
value
lane
less
Prior art date
Application number
PCT/CN2022/081190
Other languages
English (en)
French (fr)
Inventor
魏成志
何煜埕
赵志伟
李俊杰
Original Assignee
江苏航天大为科技股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 江苏航天大为科技股份有限公司 filed Critical 江苏航天大为科技股份有限公司
Publication of WO2023108932A1 publication Critical patent/WO2023108932A1/zh

Links

Images

Classifications

    • 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
    • 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
    • G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
    • 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
    • G08G1/052—Detecting movement of traffic to be counted or controlled with provision for determining speed or overspeed

Definitions

  • the invention belongs to the technical field of road traffic, and in particular relates to a method for identifying abnormal driving behavior of vehicles based on millimeter wave radar.
  • millimeter-wave radar is more and more used in vehicle detection and tracking in the field of intelligent transportation due to its all-weather work and the advantages of not being affected by weather conditions. By emitting electromagnetic waves and receiving echo signals, vehicle targets can be detected and positioned with high precision.
  • the millimeter wave radar cannot directly judge whether the vehicle has abnormal driving behaviors such as changing lanes and going backwards. As a hot spot in the field of road traffic, abnormal driving behavior contains more potential danger information, and detecting vehicles with abnormal driving behavior is crucial to safe traffic. Therefore, it is necessary to invent a method for identifying abnormal vehicle driving behavior.
  • the present invention proposes a method for identifying abnormal driving behavior of vehicles based on millimeter-wave radar.
  • This method can not only identify whether there is abnormal driving behavior in the vehicle information tracked by the millimeter-wave radar in real time, but also record the detected abnormalities. Vehicle location information, lane information, etc.
  • the purpose of the present invention is to detect abnormal driving behavior on the road based on the millimeter wave radar, including four abnormal vehicle behaviors of abnormal lane change, retrograde, abnormal speeding and emergency braking.
  • the millimeter-wave radar-based vehicle abnormal driving behavior recognition method disclosed in the present invention is applied to a vehicle abnormal driving behavior recognition system, and the system includes a vehicle real-time information acquisition unit, a lane configuration unit, a vehicle abnormal driving determination unit, and vehicle information
  • the vehicle real-time information acquisition unit acquires the vehicle real-time position and speed information collected by the millimeter-wave radar vehicle detector on the main line of the road intersection;
  • the vehicle abnormal driving determination unit judges according to the vehicle information in the vehicle storage unit and the determination algorithm Whether the vehicle has abnormal driving behavior; the horizontal and vertical cutting of the actual road conditions by the lane configuration unit;
  • the method includes: judging whether the vehicle is an incoming or outgoing vehicle according to the vehicle position and the lane it is in, and judging whether it is abnormally changing lanes or going backwards; judging whether the vehicle is an incoming or outgoing vehicle according to the vehicle position and the lane it is in, and combining the vehicle speed Judging overspeed or emergency braking.
  • the horizontal cutting cuts the actual road conditions in a direction parallel to the position of the millimeter-wave radar to form different sub-sections and At the same time, record the horizontal starting coordinate information RY m of each road section, where m is the number of the road section, and the value range is [1, 8].
  • the longitudinal cutting cuts the actual road conditions in a direction perpendicular to the position of the millimeter-wave radar to form different sub-lanes and It is used to record the lane information of the vehicle in the actual tracking process, where k is the lane number, and the value range is [1, 12].
  • the judging that the vehicle is an incoming or outgoing vehicle according to the position of the vehicle and the lane it is in includes:
  • the method for judging an abnormal lane change includes:
  • step S01 For oncoming vehicles, compare the vehicle lateral position coordinates and between the size, if less than It is determined that the road section number where the vehicle is located is m, otherwise it will be compared with Compare size, if less than It is determined that the road section number where the vehicle is located is m+1, otherwise it will be compared with Compare, and so on; if the value of the road section number where the target vehicle is located cannot be calculated in the end, set the default road section number 99, if the road section number is not equal to 99, execute step S02, otherwise judge whether the next vehicle in the vehicle information is abnormal Lane change; for the going vehicle, compare the vehicle's lateral position coordinates and between the size, if less than It is determined that the section number of the vehicle is m; otherwise, the Compare size, if less than It is determined that the road section number where the vehicle is located is m+1, otherwise it will be compared with Compare, and so on; if the value of the road section number where the target vehicle is located cannot be calculated in the end, set the default road
  • step S02 For oncoming vehicles, calculate the vehicle lateral position coordinates and The distance difference between, after taking the absolute value and Compare size, if less than Then it is judged that the temporary lane number TCL i of the current vehicle is k, otherwise it is compared with Compare size, if less than Then it is determined that the temporary lane number TCL i of the current vehicle is k+1, otherwise it will be compared with Compare, and so on, if the value of the temporary lane number TCL i of the target vehicle cannot be calculated in the end, then set the default lane number 99, if the lane number is not equal to 99, execute step S03, otherwise judge the next vehicle in the vehicle information Is there an abnormal lane change; for the going vehicle, calculate the vehicle's lateral position coordinates The distance difference between and RY m , after taking the absolute value and Compare size, if less than Then it is judged that the temporary lane number TCL i of the current vehicle is k, otherwise it is compared with Compare size, if less than
  • step S03 Check whether the road section where the target vehicle is located is a section where road changes are prohibited, if so, execute step S04, otherwise determine whether the next vehicle in the vehicle information is abnormally changing lanes;
  • TCL i For the calculated temporary lane number TCL i of the target vehicle, if TCL i is 99, the value of RCL i is not updated; otherwise, compare the value stored in RCL i , if the value in RCL i is not equal to 99, and TCL i and If the value of RCL i is not equal, add 1 to the value of temporary counter TSC i , if the value in RCL i is equal to 99, use TCL i to update the value of RCL i ;
  • S05 Compare the value of the temporary counter TSC i with the preset threshold value, if it is greater than the threshold value, it is determined that the vehicle has an abnormal lane change behavior, and update the time T i of the abnormal lane change, the target vehicle number CN i , and the target vehicle lane number CL i , use TCL i to update the value of RCL i ; otherwise, judge whether the next vehicle in the vehicle information changes lane abnormally.
  • the method for judging that the vehicle is traveling in the wrong direction includes:
  • S11 For oncoming vehicles, calculate the vehicle lateral position coordinates The distance difference between and RY m , after taking the absolute value and Compare size, if less than Then determine the current vehicle lane number is k, otherwise and Compare size, if less than Then it is determined that the current vehicle lane number is k+1, otherwise it will be compared with Compare, and so on, if the value of the lane number of the target vehicle cannot be calculated in the end, then set the default lane number 99, that is, the unknown lane number;
  • S12 Calculate the vehicle lateral position coordinates The distance difference between and RY m , after taking the absolute value and Compare size, if less than Then determine the current vehicle lane number is k, otherwise and Compare size, if less than Then it is determined that the current vehicle lane number is k+1, otherwise it will be compared with Compare, and so on, if the value of the lane number of the target vehicle cannot be calculated in the end, set the default lane number 99;
  • the method for judging abnormal speeding of the vehicle includes:
  • step S22 If the segment number of the target vehicle V i is not equal to 99, compare the longitudinal speed of the target vehicle V i If the speed limit threshold of the road section where the vehicle is located is less than the speed limit threshold, step S23 is executed, otherwise the temporary counter TSO i adds 1;
  • S24 Compare the value of the temporary counter TSO i with the given threshold, if it is less than the threshold, judge whether the next vehicle in the vehicle information is abnormally speeding; if it is equal to the threshold, judge that there is an abnormal speeding behavior, and update the target vehicle number CN i , the time T i when the abnormal speeding occurs, the lane number CL i of the target vehicle, and the duration OT i after the abnormal driving behavior of the target vehicle occurs, plus 1; if the value of TSO i is greater than the threshold, only update the occurrence of abnormal speeding behavior of the target vehicle After the duration OT i , add 1 to the value of OT i .
  • the method for judging emergency braking of the vehicle includes:
  • S31 For oncoming vehicles, calculate the vehicle lateral position coordinates The distance difference between and RY m , after taking the absolute value and Compare size, if less than Then determine the current vehicle lane number is k, otherwise and Compare size, if less than Then it is determined that the current vehicle lane number is k+1, otherwise it will be compared with Compare, and so on, if the value of the lane number of the target vehicle cannot be calculated in the end, set the default lane number 99; for the going vehicle, calculate the vehicle's lateral position coordinates The distance difference between and RY m , after taking the absolute value and Compare size, if less than Then determine the current vehicle lane number is k, otherwise and Compare size, if less than Then it is determined that the current vehicle lane number is k+1, otherwise it will be compared with Compare, and so on, if the value of the lane number of the target vehicle cannot be calculated in the end, set the default lane number 99;
  • TSEB i is equal to the interval frame number FN, update the value of RD 1 to the value of RD 2 , and set the speed of the current target vehicle V i Record to RD 2 , and then compare the values of RD 1 and RD 2 , if RD 2 is greater than RD 1 multiplied by the predetermined threshold, judge whether the next vehicle in the vehicle information is emergency braking, otherwise determine that the current target vehicle V i is in emergency For the braking behavior, update the target vehicle number CN i , the time T i of emergency braking, and the lane number CL i of the target vehicle, and clear the value of TSEB i to 0.
  • the vehicle information storage unit stores the basic information of the vehicle and abnormal driving information;
  • the basic information includes the vehicle's real-time position P i , speed information S i , and lane information RCL i ;
  • the abnormal driving information includes abnormal lane-changing behavior C i , abnormal speeding behavior O i , retrograde behavior N i , emergency braking behavior EB i , time when abnormal driving behavior occurs T i , duration of abnormal speeding behavior after occurrence OT i , duration of retrograde driving behavior after occurrence NT i , vehicle number CN i , lane number CL i where the vehicle target is located, where the speed information includes the longitudinal speed and lateral velocity Real-time position including longitudinal position coordinates and horizontal position coordinates i represents the vehicle number detected and tracked in real time, and the initial value of lane information RCL i is 99.
  • the invention stores abnormal vehicle information, provides a data basis for the analysis of traffic conditions at intersections, and simultaneously provides vehicle data information for video vehicle inspection devices to capture.
  • Fig. 1 is a flow chart of the method for judging the abnormal lane change of a vehicle in the present invention
  • Fig. 2 is a flow chart of the present invention's method for judging a vehicle traveling in reverse
  • Fig. 3 the flow chart of the present invention judging vehicle abnormal speeding method
  • Fig. 4 is a flow chart of the method for judging emergency braking of a vehicle in the present invention.
  • the present application discloses a method for identifying abnormal driving behavior of vehicles based on millimeter-wave radar.
  • Lane information achieves the effect of effectively identifying the abnormal driving behavior of the vehicle, including a vehicle real-time information acquisition unit, a vehicle information storage unit, a vehicle abnormal driving determination unit, and a lane configuration unit.
  • the vehicle abnormal driving determination unit includes four parts: detection of abnormal lane-changing vehicles, detection of retrograde vehicles, detection of abnormal speeding vehicles, and detection of emergency braking vehicles.
  • a method for identifying abnormal driving behavior of vehicles based on millimeter-wave radar As shown in Figure 1, a method for identifying abnormal driving behavior of vehicles based on millimeter-wave radar.
  • the detection steps for abnormal lane-changing vehicles are as follows:
  • the obvious feature of vehicle lane changing is that the lane number of the target vehicle V i changes. If the vehicle has abnormal lane changing behavior, the abnormal driving information in the vehicle information storage unit is related to the target vehicle’s abnormal lane changing behavior C i , vehicle number The value of CN i , the time T i when the abnormal driving behavior occurs, and the lane number CL i where the vehicle target is located will be updated, that is, there will be an information queue (C i , CN i , T i , CL i , P i ), where i represents the number of the vehicle detected and tracked in real time, and P i is the coordinate information of the target vehicle.
  • the specific detection steps are as follows:
  • Step 1 Determining whether the vehicle changes lanes abnormally is mainly based on whether the road section where the target vehicle V i is located belongs to the forbidden road section. If the road section where the target vehicle V i is located belongs to the forbidden road section, then perform step 2.
  • Step 2 Periodically collect vehicle information in the tracking system to the vehicle information storage unit.
  • Step 3 Determine the real-time position P i of the target vehicle V i in the basic information of the vehicle in the vehicle information storage unit, and compare the vehicle lateral position coordinates The size between and RY m , if If it is less than RY m , it is judged that the vehicle belongs to the vehicle on the right side of the central divider, that is, the destination vehicle. Otherwise, it is determined that the vehicle belongs to the vehicle on the left side of the central divider, that is, the oncoming vehicle.
  • Step 4 If the target vehicle V i belongs to the left vehicle, go to step 5; otherwise, go to step 6.
  • Step 5 Compare vehicle lateral position coordinates and between the size, if less than It is determined that the road segment number where the vehicle is located is m. Otherwise with Compare size, if less than It is determined that the road section number where the vehicle is located is m+1, otherwise it will be compared with compare, and so on. If the value of the road section number where the target vehicle is located cannot be calculated in the end, set the default road section number 99, that is, the unknown road section number, if the road section number is not equal to 99, go to step 7, otherwise go to step 12.
  • Step 6 Compare vehicle lateral position coordinates and between the size, if less than It is determined that the road segment number where the vehicle is located is m. Otherwise with Compare size, if less than It is determined that the road section number where the vehicle is located is m+1, otherwise it will be compared with compare, and so on. If the value of the road section number where the target vehicle is located cannot be calculated in the end, set the default road section number 99, that is, the unknown road section number, if the road section number is not equal to 99, go to step 8, otherwise go to step 12.
  • Step 7 Calculate the vehicle lateral position coordinates The distance difference between and RY m , after taking the absolute value and Compare size, if less than Then it is judged that the temporary lane number TCL i of the current vehicle is k, otherwise it is compared with Compare size, if less than Then it is determined that the temporary lane number TCL i of the current vehicle is k+1, otherwise it will be compared with Compare, and so on, if the value of the temporary lane number TCL i of the target vehicle cannot be calculated in the end, set the default lane number 99, that is, the unknown lane number, if the lane number is not equal to 99, go to step 9, otherwise go to step 12.
  • Step 8 Calculate the vehicle lateral position coordinates The distance difference between and RY m , after taking the absolute value and Compare size, if less than Then it is judged that the temporary lane number TCL i of the current vehicle is k, otherwise it is compared with Compare size, if less than Then it is determined that the temporary lane number TCL i of the current vehicle is k+1, otherwise it will be compared with Compare, and so on, if the value of the temporary lane number TCL i of the target vehicle cannot be calculated in the end, set the default lane number 99, that is, the unknown lane number, if the lane number is not equal to 99, go to step 9, otherwise go to step 12.
  • Step 9 Check whether the road segment where the target vehicle is located is a prohibited road change segment, if so, go to step 10, otherwise go to step 12.
  • Step 10 For the calculated temporary lane number TCL i of the target vehicle, if TCL i is 99, the value of RCL i is not updated. Otherwise compare the value stored in RCL i , if the value in RCL i is not equal to 99, and the values of TCL i and RCL i are not equal, add 1 to the value of temporary counter TSC i , if the value in RCL i is equal to 99, use TCL i updates the value of RCL i ;
  • Step 11 Compare the value of the temporary counter TSC i with the given threshold, if it is less than the threshold, go to step 12; otherwise, it is determined that the vehicle has an abnormal lane-changing behavior, and update the time T i of the abnormal lane-changing and the target vehicle number CN i , lane number CL i of the target vehicle, use TCL i to update the value of RCL i .
  • Step 12 traverse the next vehicle in the collected vehicle information, and perform step 3.
  • a method for identifying abnormal driving behavior of vehicles based on millimeter-wave radar has the following steps for detecting reverse vehicles:
  • the abnormal driving information in the vehicle information storage unit is related to the target vehicle’s retrograde behavior N i , vehicle number CN i , time Ti when the abnormal driving behavior occurred, and retrograde driving behavior
  • the value of the lane number CLi where the vehicle target is located will be updated for a duration of NT i after the occurrence, that is, there will be an information queue (N i , CN i , T i , NT i , CL i , P i ), where i represents The number of the tracked vehicle is detected in real time, and P i is the coordinate information of the target vehicle.
  • the specific detection steps are as follows:
  • Step 1 Periodically collect vehicle information in the tracking system to the vehicle information storage unit.
  • Step 2 Determine the real-time position P i of the target vehicle V i in the basic information of the vehicle in the vehicle information storage unit, and compare the vehicle lateral position coordinates The size between and RY m , if If it is less than RY m , it is judged that the vehicle belongs to the vehicle on the right side of the central divider, that is, the destination vehicle. Otherwise, it is determined that the vehicle belongs to the vehicle on the left side of the central divider, that is, the oncoming vehicle.
  • Step 3 If the target vehicle V i belongs to the left vehicle, go to step 4; otherwise, go to step 5.
  • Step 4 Calculate the vehicle lateral position coordinates The distance difference between and RY m , after taking the absolute value and Compare size, if less than Then determine the current vehicle lane number is k, otherwise and Compare size, if less than Then it is determined that the current vehicle lane number is k+1, otherwise it will be compared with Compare, and so on, if the value of the lane number of the target vehicle cannot be calculated in the end, then set the default lane number 99, that is, the unknown lane number.
  • Step 5 Calculate the vehicle lateral position coordinates The distance difference between and RY m , after taking the absolute value and Compare size, if less than Then determine the current vehicle lane number is k, otherwise and Compare size, if less than Then it is determined that the current vehicle lane number is k+1, otherwise it will be compared with Compare, and so on, if the value of the lane number of the target vehicle cannot be calculated in the end, then set the default lane number 99, that is, the unknown lane number.
  • Step 6 For the target vehicle V i whose lane number is not equal to 99, if the target vehicle V i belongs to the right side vehicle, and the vehicle longitudinal speed is less than 0, that is, the target vehicle on the right has the possibility of reverse driving, and the temporary counter TSN i is increased by 1; if the target vehicle V i belongs to the left vehicle, and the vehicle longitudinal speed If it is greater than 0, it means that the target vehicle on the right has the possibility of reverse driving, and the temporary counter TSN i is incremented by 1.
  • Step 7 Compare the value of the temporary counter TSN i with the given threshold value, if it is less than the threshold value, then execute step 8; if it is equal to the threshold value, then it is judged as the behavior of reverse driving, and the target vehicle number CN i and the time T of reverse driving occur are updated i , the lane number CL i of the target vehicle, add 1 to the duration NT i after the abnormal driving behavior of the target vehicle occurs; if the value of TSN i is greater than the threshold, only update the duration NT i after the abnormal driving behavior of the target vehicle occurs, and set The value of NT i is incremented by 1.
  • Step 8 Traversing the next vehicle in the collected vehicle information, go to step 2.
  • a method for identifying abnormal driving behavior of vehicles based on millimeter-wave radar As shown in Figure 3, a method for identifying abnormal driving behavior of vehicles based on millimeter-wave radar.
  • the detection steps of abnormal speeding vehicles are as follows:
  • the characteristic of the abnormal speeding vehicle is that the speed of the target vehicle exceeds the speed threshold set in a certain road section. If the vehicle has abnormal speeding behavior, the abnormal driving information in the vehicle information storage unit is about the speeding behavior O i of the target vehicle, the vehicle number The value of CN i , the time T i when the abnormal driving behavior occurs, and the lane number CLi where the vehicle target is located will be updated, that is, the information queue will be generated (O i , CN i , T i , OT i , CL i , P i ) , where i represents the number of the vehicle detected and tracked in real time, and P i is the coordinate information of the target vehicle.
  • the specific detection steps are as follows:
  • Step 1 Periodically collect vehicle information in the tracking system to the vehicle information storage unit.
  • Step 2 Determine the real-time position P i of the target vehicle V i in the basic information of the vehicle in the vehicle information storage unit, and compare the vehicle lateral position coordinates The size between and RY m , if If it is less than RY m , it is judged that the vehicle belongs to the vehicle on the right side of the central divider, that is, the destination vehicle. Otherwise, it is determined that the vehicle belongs to the vehicle on the left side of the central divider, that is, the oncoming vehicle.
  • Step 3 If the target vehicle V i belongs to the left vehicle, go to step 4; otherwise, go to step 5.
  • Step 4 Compare vehicle lateral position coordinates and between the size, if less than It is determined that the road segment number where the vehicle is located is m. Otherwise with Compare size, if less than It is determined that the road section number where the vehicle is located is m+1, otherwise it will be compared with compare, and so on. If the value of the road section number where the target vehicle is located cannot be finally calculated, the default road section number 99, which is an unknown road section number, is set. If the section number is not equal to 99, go to step 6, otherwise go to step 10.
  • Step 5 Compare vehicle lateral position coordinates and between the size, if less than It is determined that the road segment number where the vehicle is located is m. Otherwise with Compare size, if less than It is determined that the road section number where the vehicle is located is m+1, otherwise it will be compared with compare, and so on. If the value of the road section number where the target vehicle is located cannot be finally calculated, the default road section number 99, which is an unknown road section number, is set. If the segment number is not equal to 99, go to step 7, otherwise go to step 10.
  • Step 6 Calculate the vehicle lateral position coordinates The distance difference between and RY m , after taking the absolute value and Compare size, if less than Then determine the current vehicle lane number is k, otherwise and Compare size, if less than Then it is determined that the current vehicle lane number is k+1, otherwise it will be compared with Compare, and so on, if the value of the lane number of the target vehicle cannot be calculated in the end, then set the default lane number 99, that is, the unknown lane number, if the lane number is not equal to 99, go to step 8, otherwise go to step 10.
  • Step 7 Calculate the vehicle lateral position coordinates The distance difference between and RY m , after taking the absolute value and Compare size, if less than Then determine the current vehicle lane number is k, otherwise and Compare size, if less than Then it is determined that the current vehicle lane number is k+1, otherwise it will be compared with Compare, and so on, if the value of the lane number of the target vehicle cannot be calculated in the end, then set the default lane number 99, that is, the unknown lane number, if the lane number is not equal to 99, go to step 8, otherwise go to step 10.
  • Step 8 If the segment number of the target vehicle V i is not equal to 99, compare the longitudinal speed of the target vehicle V i If the speed limit threshold of the road section where the vehicle is located is less than the speed limit threshold, then execute step 10; otherwise, add 1 to the temporary counter TSO i .
  • Step 9 Compare the value of the temporary counter TSO i with the given threshold value, if it is less than the threshold value, execute step 10; if it is equal to the threshold value, then determine that there is an abnormal speeding behavior, and update the target vehicle number CN i and the time T of the abnormal speeding occurrence i , the lane number CL i of the target vehicle, add 1 to the duration OT i of the target vehicle’s abnormal driving behavior; if the value of TSO i is greater than the threshold, only update the duration of the target vehicle’s abnormal speeding behavior after the The value of OT i is incremented by 1.
  • Step 10 Traversing the next vehicle in the collected vehicle information, go to step 2.
  • a method for identifying abnormal driving behavior of vehicles based on millimeter-wave radar has the following steps for detecting emergency braking vehicles:
  • the emergency braking behavior of the vehicle is characterized by the speed of the moving vehicle suddenly dropping to a certain value, or directly entering the stop state. If the vehicle has emergency braking behavior, the abnormal driving information in the vehicle information storage unit is about the emergency braking behavior EB i of the target vehicle, the vehicle number CN i , the time Ti when the abnormal driving behavior occurs, and the lane where the vehicle target is located.
  • the value of CLi will be updated, that is, there will be an information queue (EB i , CN i , T i , CL i , P i ), where i represents the vehicle number detected and tracked in real time, and P i is the coordinate information of the target vehicle.
  • the specific detection steps are as follows:
  • Step 1 Periodically collect vehicle information in the tracking system to the vehicle information storage unit, and determine the storage interval frame number FN according to the collection cycle.
  • Step 2 Determine the real-time position P i of the target vehicle V i in the basic information of the vehicle in the vehicle information storage unit, and compare the vehicle lateral position coordinates The size between and RY m , if If it is less than RY m , it is judged that the vehicle belongs to the vehicle on the right side of the central divider, that is, the destination vehicle. Otherwise, it is determined that the vehicle belongs to the vehicle on the left side of the central divider, that is, the oncoming vehicle.
  • Step 3 If the target vehicle Vi belongs to the left vehicle, go to step 4; otherwise go to step 5.
  • Step 4 Calculate the vehicle lateral position coordinates The distance difference between and RY m , after taking the absolute value and Compare size, if less than Then determine the current vehicle lane number is k, otherwise and Compare size, if less than Then it is determined that the current vehicle lane number is k+1, otherwise it will be compared with Compare, and so on, if the value of the lane number of the target vehicle cannot be calculated in the end, set the default lane number 99, that is, the unknown lane number, if the lane number is not equal to 99, go to step 6, otherwise go to step 8.
  • Step 5 Calculate the vehicle lateral position coordinates The distance difference between and RY m , after taking the absolute value and Compare size, if less than Then determine the current vehicle lane number is k, otherwise and Compare size, if less than Then it is determined that the current vehicle lane number is k+1, otherwise it will be compared with Compare, and so on, if the value of the lane number of the target vehicle cannot be calculated in the end, set the default lane number 99, that is, the unknown lane number, if the lane number is not equal to 99, go to step 6, otherwise go to step 8.
  • Step 6 Set the value of the temporary counter TSEB i to 0, and record the speed of the target vehicle V i To RD 2 , the value of the temporary counter TSEB i is incremented by 1.
  • Step 7 If TSEB i is equal to the interval frame number FN, update the value of RD 1 to the value of RD 2 , and set the speed of the current target vehicle V i Record to RD 2 , and then compare the values of RD 1 and RD 2 , if RD 2 is greater than RD 1 multiplied by the predetermined threshold, go to step 8; otherwise, it is judged that the current target vehicle V i has an emergency braking behavior, and the target vehicle number CN is updated i , the time T i of emergency braking, the lane number CL i of the target vehicle, clear the value of TSEB i to 0;
  • Step 8 Traversing the next vehicle in the collected vehicle information, go to step 2.
  • the invention stores abnormal vehicle information, provides a data basis for the analysis of traffic conditions at intersections, and also provides vehicle data information for video vehicle inspection devices to capture
  • the word "preferred” means serving as an example, instance or illustration. Any aspect or design described herein as “preferred” is not necessarily to be construed as advantageous over other aspects or designs. Rather, use of the word “preferably” is intended to present concepts in a concrete manner.
  • the term “or” is intended to mean an inclusive “or” rather than an exclusive “or”. That is, unless otherwise specified or clear from context, "X employs A or B” is meant to naturally include either of the permutations. That is, if X employs A; X employs B; or X employs both A and B, then "X employs A or B" is satisfied in any of the foregoing instances.
  • Each functional unit in the embodiment of the present invention may be integrated into one processing module, or each unit may physically exist separately, or multiple or more of the above units may be integrated into one module.
  • the above-mentioned integrated modules can be implemented in the form of hardware or in the form of software function modules. If the integrated modules are implemented in the form of software function modules and sold or used as independent products, they can also be stored in a computer-readable storage medium.
  • the storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, and the like.
  • Each of the above devices or systems may execute the storage method in the corresponding method embodiment.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Traffic Control Systems (AREA)

Abstract

基于毫米波雷达的车辆异常行驶行为识别方法，属于道路交通领域，通过车辆实时信息获取单元获取毫米波雷达车辆检测器采集的车辆实时位置、速度信息；车辆异常行驶判定单元根据车辆存储单元中的车辆信息以及判定算法来判断车辆是否存在行驶异常行为；车道配置单元对实际路况进行横向切割与纵向切割；根据车辆位置和所处车道判断车辆为来向或去向车辆，并判断是否异常变道或逆行；根据车辆位置和所处车道判断车辆为来向或去向车辆，并结合车辆速度判断超速或紧急制动；通过实时获取车辆位置信息、速度信息，结合路口车道、路段划分，对车辆异常行驶进行判别；具有高实时性，能及时判定车辆的异常行驶行为。

Description

一种基于毫米波雷达的车辆异常行驶行为识别方法 技术领域
本发明属于道路交通技术领域，尤其涉及一种基于毫米波雷达的车辆异常行驶行为识别方法。
背景技术
目前毫米波雷达以其全天候工作、不受天气状况等优点，在智能交通领域被越来越多的应用于车辆检测跟踪。通过发射电磁波并接收回波信号，高精度对车辆目标进行探测和定位。但由于既定的交通规则，不同路段、路口实际情况都存在一定的差异性，毫米波雷达无法直接判断车辆是否存在变道、逆行等异常行驶行为。而异常行驶行为作为道路交通领域的关注热点，其包含了更多潜在危险信息，检测具有异常行驶行为的车辆对于安全交通是至关重要的。因此，有必要发明车辆异常行驶行为识别方法。
发明内容
有鉴于此，本发明提出了一种基于毫米波雷达车辆异常行驶行为识别方法，该方法既可识别处毫米波雷达实时跟踪到的车辆信息是否出现异常行驶行为，同时也能记录检测到的异常车辆的位置信息、车道信息等。
本发明基于毫米波雷达对道路上异常行驶行为进行检测为目的，包括对异常变道、逆行、异常超速、紧急制动四种车辆异常行为。
具体的，本发明公开的基于毫米波雷达的车辆异常行驶行为识别方法，应用于车辆异常行驶行为识别系统，所述系统包括车辆实时信息获取单元、车道配置单元、车辆异常行驶判定单元和车辆信息存储单元，车辆实时信息获取单元获取设置于道路路口主线路上毫米波雷达车辆检测器采集的车辆实时位置、速度信息；所述车辆异常行驶判定单元根据车辆存储单元中的车辆 信息以及判定算法来判断车辆是否存在行驶异常行为；所述车道配置单元对实际路况的横向切割与纵向切割；
所述方法包括：根据车辆位置和所处车道判断车辆为来向或去向车辆，并判断是否异常变道或逆行；根据车辆位置和所处车道判断车辆为来向或去向车辆，并结合车辆速度判断超速或紧急制动。
进一步的，所述横向切割以与毫米波雷达所处位置相平行的方向对实际路况进行切割，形成不同的子路段
Figure PCTCN2022081190-appb-000001
和
Figure PCTCN2022081190-appb-000002
同时记录每个路段的横向起始坐标信息RY m，其中m为路段编号，取值范围为[1，8]，L代表中央分隔带左侧路段，R代表中央分隔带左侧路段；
所述纵向切割以与毫米波雷达所处位置相垂直的方向对实际路况进行切割，形成不同的子车道
Figure PCTCN2022081190-appb-000003
和
Figure PCTCN2022081190-appb-000004
用于记录实际跟踪过程中的车辆所处的车道信息，其中k为车道编号，取值范围为[1，12]。
进一步的，所述根据车辆位置和所处车道判断车辆为来向或去向车辆包括：
周期采集跟踪系统中车辆信息至车辆信息存储单元；
确定车辆信息存储单元中的车辆基本信息关于目标车辆V i的实时位置P i，比较车辆横向位置坐标
Figure PCTCN2022081190-appb-000005
与RY m之间的大小，若
Figure PCTCN2022081190-appb-000006
小于RY m，判定车辆属于中央分隔带右侧车辆，即去向车辆，否则判定车辆属于中央分隔带左侧车辆，即来向车辆。
进一步的，所述判断异常变道的方法包括：
S01：对于来向车辆，比较车辆横向位置坐标
Figure PCTCN2022081190-appb-000007
与
Figure PCTCN2022081190-appb-000008
之间的大小，若
Figure PCTCN2022081190-appb-000009
小于
Figure PCTCN2022081190-appb-000010
判定车辆所在的路段号为m，否则再与
Figure PCTCN2022081190-appb-000011
比较大小，若小于
Figure PCTCN2022081190-appb-000012
判定车辆所在的路段号为m+1，否则再与
Figure PCTCN2022081190-appb-000013
比较，以此类推；若最终未能计算出目标车辆所在路段号的值，则设置默认路段号99，若路段号不等于99，执行步骤S02，否则判断车辆信息中的下一辆车是否异常变道；对于去向车辆，比较车辆横向位置坐标
Figure PCTCN2022081190-appb-000014
与
Figure PCTCN2022081190-appb-000015
之间的大小，若
Figure PCTCN2022081190-appb-000016
小于
Figure PCTCN2022081190-appb-000017
判定车辆所在的路段号为m；否则再与
Figure PCTCN2022081190-appb-000018
比较大小，若小于
Figure PCTCN2022081190-appb-000019
判定车辆所在的路段号为m+1，否则再与
Figure PCTCN2022081190-appb-000020
比较，以此类推；若最终未能计算出目标车辆所在路段号的值，则设置默认路段号99，若路段号不等于99，执行步骤S02，否则判断车辆信息中的下一辆车是否异常变道；
S02：对于来向车辆，计算车辆横向位置坐标
Figure PCTCN2022081190-appb-000021
与
Figure PCTCN2022081190-appb-000022
之间的距离差，取绝对值后与
Figure PCTCN2022081190-appb-000023
比较大小，若小于
Figure PCTCN2022081190-appb-000024
则判定当前车辆临时车道号TCL i为k，否则再与
Figure PCTCN2022081190-appb-000025
比较大小，若小于
Figure PCTCN2022081190-appb-000026
则判定当前车辆临时车道号TCL i为k+1，否则再与
Figure PCTCN2022081190-appb-000027
比较，以此类推，若最终未能计算出目标车辆临时车道号TCL i的值，则设置默认车道号99，若车道号不等于99，执行步骤S03，否则判断车辆信息中的下一辆车是否异常变道；对于去向车辆，计算车辆横向位置坐标
Figure PCTCN2022081190-appb-000028
与RY m之间的距离差，取绝对值后与
Figure PCTCN2022081190-appb-000029
比较大小，若小于
Figure PCTCN2022081190-appb-000030
则判定当前车辆临时车道号TCL i为k，否则再与
Figure PCTCN2022081190-appb-000031
比较大小，若小于
Figure PCTCN2022081190-appb-000032
则判定当前车辆临时车道号TCL i为k+1，否则再与
Figure PCTCN2022081190-appb-000033
比较，以此类推，若最终未能计算出目标车辆临时车道号TCL i的值，则设置默认车道号99，若车道号不等于99，执行步骤S03，否则判断车辆信息中的下一辆车是否异常变道；
S03：查看目标车辆所在的路段是否为禁止变道路段，若是，执行步骤S04，否则判断车辆信息中的下一辆车是否异常变道；
S04：对计算出的目标车辆临时车道号TCL i，若TCL i为99，则不更新RCL i值；否则对比RCL i中存储的值，若RCL i中的值不等于99，且TCL i与RCL i 的值不等，则临时计数器TSC i值加1，若RCL i中的值等于99，则使用TCL i更新RCL i的值；
S05：比较临时计数器TSC i的值与预设阈值的大小，若大于阈值，判定该车辆存在异常变道行驶行为，更新异常变道发生的时间T i、目标车辆编号CN i、目标车辆车道号CL i，使用TCL i更新RCL i的值；否则判断车辆信息中的下一辆车是否异常变道。
进一步的，所述判断车辆逆行的方法包括：
S11：对于来向车辆，计算车辆横向位置坐标
Figure PCTCN2022081190-appb-000034
与RY m之间的距离差，取绝对值后与
Figure PCTCN2022081190-appb-000035
比较大小，若小于
Figure PCTCN2022081190-appb-000036
则判定当前车辆车道号为k，否则再与
Figure PCTCN2022081190-appb-000037
比较大小，若小于
Figure PCTCN2022081190-appb-000038
则判定当前车辆车道号为k+1，否则再与
Figure PCTCN2022081190-appb-000039
比较，以此类推，若最终未能计算出目标车辆车道号的值，则设置默认车道号99，即未知车道号；
S12：计算车辆横向位置坐标
Figure PCTCN2022081190-appb-000040
与RY m之间的距离差，取绝对值后与
Figure PCTCN2022081190-appb-000041
比较大小，若小于
Figure PCTCN2022081190-appb-000042
则判定当前车辆车道号为k，否则再与
Figure PCTCN2022081190-appb-000043
比较大小，若小于
Figure PCTCN2022081190-appb-000044
则判定当前车辆车道号为k+1，否则再与
Figure PCTCN2022081190-appb-000045
比较，以此类推，若最终未能计算出目标车辆车道号的值，则设置默认车道号99；
S13：对于车道号不等于99的目标车辆V i，若目标车辆V i属于右侧车辆，并且车辆纵向速度
Figure PCTCN2022081190-appb-000046
小于0，即右侧目标车辆存在逆向行驶可能性，临时计数器TSN i加1；若目标车辆V i属于左侧车辆，并且车辆纵向速度
Figure PCTCN2022081190-appb-000047
大于0，即右侧目标车辆存在逆向行驶可能性，临时计数器TSN i加1；
S14：比较临时计数器TSN i的值与给定阈值的大小，若等于阈值，则判定为逆向行驶行为，更新目标车辆编号CN i、逆向行驶发生的时间T i、目标车 辆车道号CL i，将目标车辆异常行驶行为发生后持续的时长NT i加1；若TSN i的值大于阈值，则只更新目标车辆异常行驶行为发生后持续的时长NT i，将NT i的值加1；若小于阈值，则判断车辆信息中的下一辆车是否逆行。
进一步的，所述判断车辆异常超速的方法包括：
S21：对于来向车辆，比较车辆横向位置坐标
Figure PCTCN2022081190-appb-000048
与
Figure PCTCN2022081190-appb-000049
之间的大小，若
Figure PCTCN2022081190-appb-000050
小于
Figure PCTCN2022081190-appb-000051
判定车辆所在的路段号为m；否则再与
Figure PCTCN2022081190-appb-000052
比较大小，若小于
Figure PCTCN2022081190-appb-000053
判定车辆所在的路段号为m+1，否则再与
Figure PCTCN2022081190-appb-000054
比较，以此类推，若最终未能计算出目标车辆所在路段号的值，则设置默认路段号99，即未知路段号；对于去向车辆，比较车辆横向位置坐标
Figure PCTCN2022081190-appb-000055
与
Figure PCTCN2022081190-appb-000056
之间的大小，若
Figure PCTCN2022081190-appb-000057
小于
Figure PCTCN2022081190-appb-000058
判定车辆所在的路段号为m；否则再与
Figure PCTCN2022081190-appb-000059
比较大小，若小于
Figure PCTCN2022081190-appb-000060
判定车辆所在的路段号为m+1，否则再与
Figure PCTCN2022081190-appb-000061
比较，以此类推；若最终未能计算出目标车辆所在路段号的值，则设置默认路段号99；
S22：若目标车辆V i的的路段号不等于99，比较目标车辆V i的纵向速度
Figure PCTCN2022081190-appb-000062
与该车辆所在的路段的限速阈值，若小于限速阈值，则执行步骤S23，否则临时计数器TSO i加1；
S23：对于来向车辆，计算车辆横向位置坐标
Figure PCTCN2022081190-appb-000063
与RY m之间的距离差，取绝对值后与
Figure PCTCN2022081190-appb-000064
比较大小，若小于
Figure PCTCN2022081190-appb-000065
则判定当前车辆车道号为k，否则再与
Figure PCTCN2022081190-appb-000066
比较大小，若小于
Figure PCTCN2022081190-appb-000067
则判定当前车辆车道号为k+1，否则再与
Figure PCTCN2022081190-appb-000068
比较，以此类推，若最终未能计算出目标车辆车道号的值，则设置默认车道号99，即未知车道号；对于去向车辆，计算车辆横向位置坐标
Figure PCTCN2022081190-appb-000069
与RY m之间的距离差，取绝对值后与
Figure PCTCN2022081190-appb-000070
比较大小，若小于
Figure PCTCN2022081190-appb-000071
则判定当前车辆车道号为k，否则再与
Figure PCTCN2022081190-appb-000072
比较大小，若小于
Figure PCTCN2022081190-appb-000073
则判定当前车辆车道号为k+1，否则再与
Figure PCTCN2022081190-appb-000074
比较，以此类推，若最终未能计算出目标车辆车道号的值，则设置默认车道号99；
S24：比较临时计数器TSO i的值与给定阈值的大小，若小于阈值，则判断车辆信息中的下一辆车是否异常超速；若等于阈值，则判定存在异常超速行为，更新目标车辆编号CN i、异常超速发生的时间T i，目标车辆的车道号CL i，目标车辆异常行驶行为发生后持续的时长OT i加1；若TSO i的值大于阈值，则只更新目标车辆异常超速行为发生后持续的时长OT i，将OT i的值加1。
进一步的，所述判断车辆紧急制动的方法包括：
S31：对于来向车辆，计算车辆横向位置坐标
Figure PCTCN2022081190-appb-000075
与RY m之间的距离差，取绝对值后与
Figure PCTCN2022081190-appb-000076
比较大小，若小于
Figure PCTCN2022081190-appb-000077
则判定当前车辆车道号为k，否则再与
Figure PCTCN2022081190-appb-000078
比较大小，若小于
Figure PCTCN2022081190-appb-000079
则判定当前车辆车道号为k+1，否则再与
Figure PCTCN2022081190-appb-000080
比较，以此类推，若最终未能计算出目标车辆车道号的值，则设置默认车道号99；对于去向车辆，计算车辆横向位置坐标
Figure PCTCN2022081190-appb-000081
与RY m之间的距离差，取绝对值后与
Figure PCTCN2022081190-appb-000082
比较大小，若小于
Figure PCTCN2022081190-appb-000083
则判定当前车辆车道号为k，否则再与
Figure PCTCN2022081190-appb-000084
比较大小，若小于
Figure PCTCN2022081190-appb-000085
则判定当前车辆车道号为k+1，否则再与
Figure PCTCN2022081190-appb-000086
比较，以此类推，若最终未能计算出目标车辆车道号的值，则设置默认车道号99；
S32：设置临时计数器TSEB i值为0，记录目标车辆V i的速度
Figure PCTCN2022081190-appb-000087
至RD 2，临时计数器TSEB i的值加1；
S33：若TSEB i等于间隔帧数FN，则将RD 1的值更新为RD 2的值，将当前目标车辆V i的速度
Figure PCTCN2022081190-appb-000088
记录至RD 2，再比较RD 1和RD 2的值，若满足RD 2大于RD 1乘以预定阈值，判断车辆信息中的下一辆车是否紧急制动，否则判定当前目标车辆V i存在紧急制动行驶行为，更新目标车辆编号CN i、紧急制动发生 的时间T i、目标车辆的车道号CL i，将TSEB i的值清0。
进一步的，所述车辆信息存储单元存储车辆的基本信息以及异常行驶信息；所述基本信息包括车辆实时位置P i、速度信息S i，车道信息RCL i；所述异常行驶信息包括异常变道行为C i、异常超速行为O i、逆行行为N i、紧急制动行为EB i，异常行驶行为发生的时间T i，异常超速行驶行为发生后持续的时长OT i，逆行行驶行为发生后持续的时长NT i，车辆编号CN i，车辆目标所处的车道号CL i，其中速度信息包括纵向速度
Figure PCTCN2022081190-appb-000089
和横向速度
Figure PCTCN2022081190-appb-000090
实时位置包括纵向位置坐标
Figure PCTCN2022081190-appb-000091
和横向位置坐标
Figure PCTCN2022081190-appb-000092
i表示实时检测跟踪到的车辆编号，车道信息RCL i初始值为99。
本发明的有益效果如下：
通过实时获取车辆位置信息、速度信息，结合路口车道、路段划分，对车辆异常行驶进行判别；应用场景广泛，对于不同路况的道路信息，只需配置对应的车道、路段信息，弥补毫米波雷达不能直接判定车辆是否存在异常行驶行为的缺点；
具有高实时性，能及时判定车辆的异常行驶行为；
本发明存储了异常车辆信息，为路口交通情况分析提供数据基础，同时也为视频车检器抓拍提供车辆数据信息。
附图说明
图1本发明判断车辆异常变道方法的流程图；
图2本发明判断车辆逆行方法的流程图；
图3本发明判断车辆异常超速方法的流程图；
图4本发明判断车辆紧急制动方法的流程图。
具体实施方式
下面结合附图对本发明作进一步的说明，但不以任何方式对本发明加以限制，基于本发明教导所作的任何变换或替换，均属于本发明的保护范围。
参见如图1至图4所示，本申请公开了一种基于毫米波雷达车辆异常行驶行为识别方法，该方法基于毫米波雷达实现，通过实时获取检测跟踪的车辆信息，结合预配置的路段、车道信息，达到对车辆的异常行驶行为进行有效识别的效果，包含车辆实时信息获取单元、车辆信息存储单元、车辆异常行驶判定单元、车道配置单元。车辆异常行驶判定单元包含了对异常变道车辆的检测、对逆行车辆的检测、对异常超速车辆的检测、对紧急制动车辆的检测四个部分。
(1)对异常变道车辆的检测
如图1所示，一种基于毫米波雷达车辆异常行驶行为识别方法关于异常变道车辆的检测步骤如下：
车辆变道的明显特征在于目标车辆V i的车道号发生变化，若车辆存在异常变道行驶行为，则车辆信息存储单元中的异常行驶信息关于该目标车辆的异常变道行为C i、车辆编号CN i、异常行驶行为发生的时间T i、车辆目标所处的车道号CL i的值将被更新，即会存在信息队列(C i，CN i，T i，CL i，P i)，其中i表示实时检测跟踪到的车辆编号，P i为目标车辆坐标信息。具体检测步骤如下：
步骤1：判定车辆是否异常变道主要是依据目标车辆V i所在的路段是否属于禁止变道路段，若目标车辆V i所在的路段属于禁止变道路段，则执行步骤2。
步骤2：周期采集跟踪系统中车辆信息至车辆信息存储单元。
步骤3：确定车辆信息存储单元中的车辆基本信息关于该目标车辆V i的实 时位置P i，比较车辆横向位置坐标
Figure PCTCN2022081190-appb-000093
与RY m之间的大小，若
Figure PCTCN2022081190-appb-000094
小于RY m，判定车辆属于中央分隔带右侧车辆，即去向车辆。否则判定车辆属于中央分隔带左侧车辆，即来向车辆。
步骤4：若该目标车辆V i属于左侧车辆，执行步骤5；否则执行步骤6。
步骤5：比较车辆横向位置坐标
Figure PCTCN2022081190-appb-000095
与
Figure PCTCN2022081190-appb-000096
之间的大小，若
Figure PCTCN2022081190-appb-000097
小于
Figure PCTCN2022081190-appb-000098
判定车辆所在的路段号为m。否则再与
Figure PCTCN2022081190-appb-000099
比较大小，若小于
Figure PCTCN2022081190-appb-000100
判定车辆所在的路段号为m+1，否则再与
Figure PCTCN2022081190-appb-000101
比较，以此类推。若最终未能计算出目标车辆所在路段号的值，则设置默认路段号99，即未知路段号，若路段号不等于99，执行步骤7，否则执行步骤12。
步骤6：比较车辆横向位置坐标
Figure PCTCN2022081190-appb-000102
与
Figure PCTCN2022081190-appb-000103
之间的大小，若
Figure PCTCN2022081190-appb-000104
小于
Figure PCTCN2022081190-appb-000105
判定车辆所在的路段号为m。否则再与
Figure PCTCN2022081190-appb-000106
比较大小，若小于
Figure PCTCN2022081190-appb-000107
判定车辆所在的路段号为m+1，否则再与
Figure PCTCN2022081190-appb-000108
比较，以此类推。若最终未能计算出目标车辆所在路段号的值，则设置默认路段号99，即未知路段号，若路段号不等于99，执行步骤8，否则执行步骤12。
步骤7：计算车辆横向位置坐标
Figure PCTCN2022081190-appb-000109
与RY m之间的距离差，取绝对值后与
Figure PCTCN2022081190-appb-000110
比较大小，若小于
Figure PCTCN2022081190-appb-000111
则判定当前车辆临时车道号TCL i为k，否则再与
Figure PCTCN2022081190-appb-000112
比较大小，若小于
Figure PCTCN2022081190-appb-000113
则判定当前车辆临时车道号TCL i为k+1，否则再与
Figure PCTCN2022081190-appb-000114
比较，以此类推，若最终未能计算出目标车辆临时车道号TCL i的值，则设置默认车道号99，即未知车道号，若车道号不等于99，执行步骤9，否则执行步骤12。
步骤8：计算车辆横向位置坐标
Figure PCTCN2022081190-appb-000115
与RY m之间的距离差，取绝对值后与
Figure PCTCN2022081190-appb-000116
比较大小，若小于
Figure PCTCN2022081190-appb-000117
则判定当前车辆临时车道号TCL i为k，否则再与
Figure PCTCN2022081190-appb-000118
比较大小，若小于
Figure PCTCN2022081190-appb-000119
则判定当前车辆临时车道号TCL i为k+1，否则再与
Figure PCTCN2022081190-appb-000120
比 较，以此类推，若最终未能计算出目标车辆临时车道号TCL i的值，则设置默认车道号99，即未知车道号，若车道号不等于99，执行步骤9，否则执行步骤12。
步骤9：查看目标车辆所在的路段是否为禁止变道路段，若是，执行步骤10，否则执行步骤12。
步骤10：对计算出的目标车辆临时车道号TCL i，若TCL i为99，则不更新RCL i值。否则对比RCL i中存储的值，若RCL i中的值不等于99，且TCL i与RCL i的值不等，则临时计数器TSC i值加1，若RCL i中的值等于99，则使用TCL i更新RCL i的值；
步骤11：比较临时计数器TSC i的值与给定阈值的大小，若小于阈值，则执行步骤12；否则判定该车辆存在异常变道行驶行为，更新异常变道发生的时间T i、目标车辆编号CN i、目标车辆车道号CL i，使用TCL i更新RCL i的值。
步骤12:遍历采集的车辆信息中的下一辆车，执行步骤3。
(2)对逆行车辆的检测
如图2所示，一种基于毫米波雷达车辆异常行驶行为识别方法关于逆行车辆的检测步骤如下：
对划分的所有子路段
Figure PCTCN2022081190-appb-000121
都进行车辆逆行行为检测，若车辆存在逆向行驶行为，则车辆信息存储单元中的异常行驶信息关于该目标车辆的逆行行为N i、车辆编号CN i、异常行驶行为发生的时间Ti、逆行行驶行为发生后持续的时长NT i，车辆目标所处的车道号CLi的值将被更新，即会存在信息队列(N i，CN i，T i，NT i，CL i，P i)，其中i表示实时检测跟踪到的车辆编号，P i为目标车辆坐标信息。具体检测步骤如下：
步骤1：周期采集跟踪系统中车辆信息至车辆信息存储单元。
步骤2：确定车辆信息存储单元中的车辆基本信息关于该目标车辆V i的实时位置P i，比较车辆横向位置坐标
Figure PCTCN2022081190-appb-000122
与RY m之间的大小，若
Figure PCTCN2022081190-appb-000123
小于RY m，判定车辆属于中央分隔带右侧车辆，即去向车辆。否则判定车辆属于中央分隔带左侧车辆，即来向车辆。
步骤3：若该目标车辆V i属于左侧车辆，执行步骤4；否则执行步骤5。
步骤4：计算车辆横向位置坐标
Figure PCTCN2022081190-appb-000124
与RY m之间的距离差，取绝对值后与
Figure PCTCN2022081190-appb-000125
比较大小，若小于
Figure PCTCN2022081190-appb-000126
则判定当前车辆车道号为k，否则再与
Figure PCTCN2022081190-appb-000127
比较大小，若小于
Figure PCTCN2022081190-appb-000128
则判定当前车辆车道号为k+1，否则再与
Figure PCTCN2022081190-appb-000129
比较，以此类推，若最终未能计算出目标车辆车道号的值，则设置默认车道号99，即未知车道号。
步骤5：计算车辆横向位置坐标
Figure PCTCN2022081190-appb-000130
与RY m之间的距离差，取绝对值后与
Figure PCTCN2022081190-appb-000131
比较大小，若小于
Figure PCTCN2022081190-appb-000132
则判定当前车辆车道号为k，否则再与
Figure PCTCN2022081190-appb-000133
比较大小，若小于
Figure PCTCN2022081190-appb-000134
则判定当前车辆车道号为k+1，否则再与
Figure PCTCN2022081190-appb-000135
比较，以此类推，若最终未能计算出目标车辆车道号的值，则设置默认车道号99，即未知车道号。
步骤6：对于车道号不等于99的目标车辆V i，若目标车辆V i属于右侧车辆，并且车辆纵向速度
Figure PCTCN2022081190-appb-000136
小于0，即右侧目标车辆存在逆向行驶可能性，临时计数器TSN i加1；若目标车辆V i属于左侧车辆，并且车辆纵向速度
Figure PCTCN2022081190-appb-000137
大于0，即右侧目标车辆存在逆向行驶可能性，临时计数器TSN i加1。
步骤7：比较临时计数器TSN i的值与给定阈值的大小，若小于阈值，则执行步骤8；若等于阈值，则判定为逆向行驶行为，更新目标车辆编号CN i、逆向行驶发生的时间T i、目标车辆车道号CL i，将目标车辆异常行驶行为发生后持续的时长NT i加1；若TSN i的值大于阈值，则只更新目标车辆异常行驶行为发生后持续的时长NT i，将NT i的值加1。
步骤8：遍历采集的车辆信息中的下一辆车，执行步骤2。
(3)对异常超速车辆的检测
如图3所示，一种基于毫米波雷达车辆异常行驶行为识别方法关于异常超速车辆的检测步骤如下：
异常超速车辆的特征在于目标车辆的速度超过了某个路段设置的速度阈值，若车辆存在异常超速行驶行为，则车辆信息存储单元中的异常行驶信息关于该目标车辆的超速行为O i、车辆编号CN i、异常行驶行为发生的时间T i、车辆目标所处的车道号CLi的值将被更新，即会产生信息队列(O i，CN i，T i，OT i，CL i，P i)，其中i表示实时检测跟踪到的车辆编号，P i为目标车辆坐标信息。具体检测步骤如下：
步骤1：周期采集跟踪系统中车辆信息至车辆信息存储单元。
步骤2：确定车辆信息存储单元中的车辆基本信息关于该目标车辆V i的实时位置P i，比较车辆横向位置坐标
Figure PCTCN2022081190-appb-000138
与RY m之间的大小，若
Figure PCTCN2022081190-appb-000139
小于RY m，判定车辆属于中央分隔带右侧车辆，即去向车辆。否则判定车辆属于中央分隔带左侧车辆，即来向车辆。
步骤3：若该目标车辆V i属于左侧车辆，执行步骤4；否则执行步骤5。
步骤4：比较车辆横向位置坐标
Figure PCTCN2022081190-appb-000140
与
Figure PCTCN2022081190-appb-000141
之间的大小，若
Figure PCTCN2022081190-appb-000142
小于
Figure PCTCN2022081190-appb-000143
判定车辆所在的路段号为m。否则再与
Figure PCTCN2022081190-appb-000144
比较大小，若小于
Figure PCTCN2022081190-appb-000145
判定车辆所在的路段号为m+1，否则再与
Figure PCTCN2022081190-appb-000146
比较，以此类推。若最终未能计算出目标车辆所在路段号的值，则设置默认路段号99，即未知路段号。若路段号不等于99，执行步骤6，否则执行步骤10。
步骤5：比较车辆横向位置坐标
Figure PCTCN2022081190-appb-000147
与
Figure PCTCN2022081190-appb-000148
之间的大小，若
Figure PCTCN2022081190-appb-000149
小于
Figure PCTCN2022081190-appb-000150
判定车辆所在的路段号为m。否则再与
Figure PCTCN2022081190-appb-000151
比较大小，若小于
Figure PCTCN2022081190-appb-000152
判定车辆所在的路段号为m+1，否则再与
Figure PCTCN2022081190-appb-000153
比较，以此类推。若最终未能计算 出目标车辆所在路段号的值，则设置默认路段号99，即未知路段号。若路段号不等于99，执行步骤7，否则执行步骤10。
步骤6：计算车辆横向位置坐标
Figure PCTCN2022081190-appb-000154
与RY m之间的距离差，取绝对值后与
Figure PCTCN2022081190-appb-000155
比较大小，若小于
Figure PCTCN2022081190-appb-000156
则判定当前车辆车道号为k，否则再与
Figure PCTCN2022081190-appb-000157
比较大小，若小于
Figure PCTCN2022081190-appb-000158
则判定当前车辆车道号为k+1，否则再与
Figure PCTCN2022081190-appb-000159
比较，以此类推，若最终未能计算出目标车辆车道号的值，则设置默认车道号99，即未知车道号，若车道号不等于99，执行步骤8，否则执行步骤10。
步骤7：计算车辆横向位置坐标
Figure PCTCN2022081190-appb-000160
与RY m之间的距离差，取绝对值后与
Figure PCTCN2022081190-appb-000161
比较大小，若小于
Figure PCTCN2022081190-appb-000162
则判定当前车辆车道号为k，否则再与
Figure PCTCN2022081190-appb-000163
比较大小，若小于
Figure PCTCN2022081190-appb-000164
则判定当前车辆车道号为k+1，否则再与
Figure PCTCN2022081190-appb-000165
比较，以此类推，若最终未能计算出目标车辆车道号的值，则设置默认车道号99，即未知车道号，若车道号不等于99，执行步骤8，否则执行步骤10。
步骤8：若目标车辆V i的的路段号不等于99，比较目标车辆V i的纵向速度
Figure PCTCN2022081190-appb-000166
与该车辆所在的路段的限速阈值，若小于限速阈值，则执行步骤10，否则临时计数器TSO i加1。
步骤9：比较临时计数器TSO i的值与给定阈值的大小，若小于阈值，则执行步骤10；若等于阈值，则判定存在异常超速行为，更新目标车辆编号CN i、异常超速发生的时间T i，目标车辆的车道号CL i，目标车辆异常行驶行为发生后持续的时长OT i加1；若TSO i的值大于阈值，则只更新目标车辆异常超速行为发生后持续的时长OT i，将OT i的值加1。
步骤10：遍历采集的车辆信息中的下一辆车，执行步骤2。
(4)对紧急制动车辆的检测
如图4所示，一种基于毫米波雷达车辆异常行驶行为识别方法关于紧急制 动车辆的检测步骤如下：
车辆发生紧急制动行为的特征为处于运动中车辆的速度骤降至某一值，或直接进入停止状态。若车辆存在紧急制动行驶行为，则车辆信息存储单元中的异常行驶信息关于该目标车辆的紧急制动行为EB i、车辆编号CN i、异常行驶行为发生的时间Ti、车辆目标所处的车道号CLi的值将被更新，即会存在信息队列(EB i，CN i，T i，CL i，P i)，其中i表示实时检测跟踪到的车辆编号，P i为目标车辆坐标信息。具体检测步骤如下：
步骤1：周期采集跟踪系统中车辆信息至车辆信息存储单元，依据采集周期确定存储间隔帧数FN。
步骤2：确定车辆信息存储单元中的车辆基本信息关于该目标车辆V i的实时位置P i，比较车辆横向位置坐标
Figure PCTCN2022081190-appb-000167
与RY m之间的大小，若
Figure PCTCN2022081190-appb-000168
小于RY m，判定车辆属于中央分隔带右侧车辆，即去向车辆。否则判定车辆属于中央分隔带左侧车辆，即来向车辆。
步骤3：若该目标车辆Vi属于左侧车辆，执行步骤4；否则执行步骤5。
步骤4：计算车辆横向位置坐标
Figure PCTCN2022081190-appb-000169
与RY m之间的距离差，取绝对值后与
Figure PCTCN2022081190-appb-000170
比较大小，若小于
Figure PCTCN2022081190-appb-000171
则判定当前车辆车道号为k，否则再与
Figure PCTCN2022081190-appb-000172
比较大小，若小于
Figure PCTCN2022081190-appb-000173
则判定当前车辆车道号为k+1，否则再与
Figure PCTCN2022081190-appb-000174
比较，以此类推，若最终未能计算出目标车辆车道号的值，则设置默认车道号99，即未知车道号，若车道号不等于99，执行步骤6，否则执行步骤8。
步骤5：计算车辆横向位置坐标
Figure PCTCN2022081190-appb-000175
与RY m之间的距离差，取绝对值后与
Figure PCTCN2022081190-appb-000176
比较大小，若小于
Figure PCTCN2022081190-appb-000177
则判定当前车辆车道号为k，否则再与
Figure PCTCN2022081190-appb-000178
比较大小，若小于
Figure PCTCN2022081190-appb-000179
则判定当前车辆车道号为k+1，否则再与
Figure PCTCN2022081190-appb-000180
比较，以此类推，若最终未能计算出目标车辆车道号的值，则设置默认车道号99，即未知车道号， 若车道号不等于99，执行步骤6，否则执行步骤8。
步骤6：设置临时计数器TSEB i值为0，记录目标车辆V i的速度
Figure PCTCN2022081190-appb-000181
至RD 2，临时计数器TSEB i的值加1。
步骤7：若TSEB i等于间隔帧数FN，则将RD 1的值更新为RD 2的值，将当前目标车辆V i的速度
Figure PCTCN2022081190-appb-000182
记录至RD 2，再比较RD 1和RD 2的值，若满足RD 2大于RD 1乘以预定阈值，执行步骤8，否则判定当前目标车辆V i存在紧急制动行驶行为，更新目标车辆编号CN i、紧急制动发生的时间T i、目标车辆的车道号CL i，将TSEB i的值清0；
步骤8：遍历采集的车辆信息中的下一辆车，执行步骤2。
本发明的有益效果如下：
通过实时获取车辆位置信息、速度信息，结合路口车道、路段划分，对车辆异常行驶进行判别；应用场景广泛，对于不同路况的道路信息，只需配置对应的车道、路段信息，弥补毫米波雷达不能直接判定车辆是否存在异常行驶行为的缺点；
具有高实时性，能及时判定车辆的异常行驶行为；
本发明存储了异常车辆信息，为路口交通情况分析提供数据基础，同时也为视频车检器抓拍提供车辆数据信息
本文所使用的词语“优选的”意指用作实例、示例或例证。本文描述为“优选的”任意方面或设计不必被解释为比其他方面或设计更有利。相反，词语“优选的”的使用旨在以具体方式提出概念。如本申请中所使用的术语“或”旨在意指包含的“或”而非排除的“或”。即，除非另外指定或从上下文中清楚，“X使用A或B”意指自然包括排列的任意一个。即，如果X使用A；X使用B；或X使用A和B二者，则“X使用A或B”在前述任一示例中得到满足。
而且，尽管已经相对于一个或实现方式示出并描述了本公开，但是本领域技术人员基于对本说明书和附图的阅读和理解将会想到等价变型和修改。本公开包括所有这样的修改和变型，并且仅由所附权利要求的范围限制。特别地关于由上述组件(例如元件等)执行的各种功能，用于描述这样的组件的术语旨在对应于执行所述组件的指定功能(例如其在功能上是等价的)的任意组件(除非另外指示)，即使在结构上与执行本文所示的本公开的示范性实现方式中的功能的公开结构不等同。此外，尽管本公开的特定特征已经相对于若干实现方式中的仅一个被公开，但是这种特征可以与如可以对给定或特定应用而言是期望和有利的其他实现方式的一个或其他特征组合。而且，就术语“包括”、“具有”、“含有”或其变形被用在具体实施方式或权利要求中而言，这样的术语旨在以与术语“包含”相似的方式包括。
本发明实施例中的各功能单元可以集成在一个处理模块中，也可以是各个单元单独物理存在，也可以多个或多个以上单元集成在一个模块中。上述集成的模块既可以采用硬件的形式实现，也可以采用软件功能模块的形式实现。所述集成的模块如果以软件功能模块的形式实现并作为独立的产品销售或使用时，也可以存储在一个计算机可读取存储介质中。上述提到的存储介质可以是只读存储器，磁盘或光盘等。上述的各装置或系统，可以执行相应方法实施例中的存储方法。
综上所述，上述实施例为本发明的一种实施方式，但本发明的实施方式并不受所述实施例的限制，其他的任何背离本发明的精神实质与原理下所做的改变、修饰、代替、组合、简化，均应为等效的置换方式，都包含在本发明的保护范围之内。

Claims (8)

  1. 一种基于毫米波雷达的车辆异常行驶行为识别方法，应用于车辆异常行驶行为识别系统，其特征在于，所述系统包括车辆实时信息获取单元、车道配置单元、车辆异常行驶判定单元和车辆信息存储单元，车辆实时信息获取单元获取设置于道路路口主线路上毫米波雷达车辆检测器采集的车辆实时位置、速度信息；所述车辆异常行驶判定单元根据车辆存储单元中的车辆信息以及判定算法来判断车辆是否存在行驶异常行为；所述车道配置单元对实际路况的横向切割与纵向切割；
    所述方法包括：根据车辆位置和所处车道判断车辆为来向或去向车辆，并判断是否异常变道或逆行；根据车辆位置和所处车道判断车辆为来向或去向车辆，并结合车辆速度判断超速或紧急制动。
  2. 根据权利要求1所述的一种基于毫米波雷达的车辆异常行驶行为识别方法，其特征在于，所述横向切割以与毫米波雷达所处位置相平行的方向对实际路况进行切割，形成不同的子路段
    Figure PCTCN2022081190-appb-100001
    和
    Figure PCTCN2022081190-appb-100002
    同时记录每个路段的横向起始坐标信息RY m，其中m为路段编号，取值范围为[1，8]，L代表中央分隔带左侧路段，R代表中央分隔带左侧路段；
    所述纵向切割以与毫米波雷达所处位置相垂直的方向对实际路况进行切割，形成不同的子车道
    Figure PCTCN2022081190-appb-100003
    和
    Figure PCTCN2022081190-appb-100004
    用于记录实际跟踪过程中的车辆所处的车道信息，其中k为车道编号，取值范围为[1，12]。
  3. 根据权利要求2所述的一种基于毫米波雷达的车辆异常行驶行为识别方法，其特征在于，所述根据车辆位置和所处车道判断车辆为来向或去向车辆包括：
    周期采集跟踪系统中车辆信息至车辆信息存储单元；
    确定车辆信息存储单元中的车辆基本信息关于目标车辆V i的实时位置P i，比较车辆横向位置坐标
    Figure PCTCN2022081190-appb-100005
    与RY m之间的大小，若
    Figure PCTCN2022081190-appb-100006
    小于RY m，判定 车辆属于中央分隔带右侧车辆，即去向车辆，否则判定车辆属于中央分隔带左侧车辆，即来向车辆。
  4. 根据权利要求3所述的一种基于毫米波雷达的车辆异常行驶行为识别方法，其特征在于，所述判断异常变道的方法包括：
    S01：对于来向车辆，比较车辆横向位置坐标
    Figure PCTCN2022081190-appb-100007
    与
    Figure PCTCN2022081190-appb-100008
    之间的大小，若
    Figure PCTCN2022081190-appb-100009
    小于
    Figure PCTCN2022081190-appb-100010
    判定车辆所在的路段号为m，否则再与
    Figure PCTCN2022081190-appb-100011
    比较大小，若小于
    Figure PCTCN2022081190-appb-100012
    判定车辆所在的路段号为m+1，否则再与
    Figure PCTCN2022081190-appb-100013
    比较，以此类推；对于去向车辆，比较车辆横向位置坐标
    Figure PCTCN2022081190-appb-100014
    与
    Figure PCTCN2022081190-appb-100015
    之间的大小，若
    Figure PCTCN2022081190-appb-100016
    小于
    Figure PCTCN2022081190-appb-100017
    判定车辆所在的路段号为m；否则再与
    Figure PCTCN2022081190-appb-100018
    比较大小，若小于
    Figure PCTCN2022081190-appb-100019
    判定车辆所在的路段号为m+1，否则再与
    Figure PCTCN2022081190-appb-100020
    比较，以此类推；若最终未能计算出来向或去向车辆所在路段号的值，则设置默认路段号99，若路段号不等于99，执行步骤S02，否则判断车辆信息中的下一辆车是否异常变道；
    S02：对于来向车辆，计算车辆横向位置坐标
    Figure PCTCN2022081190-appb-100021
    与RY m之间的距离差，取绝对值后与
    Figure PCTCN2022081190-appb-100022
    比较大小，若小于
    Figure PCTCN2022081190-appb-100023
    则判定当前车辆临时车道号TCL i为k，否则再与
    Figure PCTCN2022081190-appb-100024
    比较大小，若小于
    Figure PCTCN2022081190-appb-100025
    则判定当前车辆临时车道号TCL i为k+1，否则再与
    Figure PCTCN2022081190-appb-100026
    比较，以此类推；对于去向车辆，计算车辆横向位置坐标
    Figure PCTCN2022081190-appb-100027
    与RY m之间的距离差，取绝对值后与
    Figure PCTCN2022081190-appb-100028
    比较大小，若小于
    Figure PCTCN2022081190-appb-100029
    则判定当前车辆临时车道号TCL i为k，否则再与
    Figure PCTCN2022081190-appb-100030
    比较大小，若小于
    Figure PCTCN2022081190-appb-100031
    则判定当前车辆临时车道号TCL i为k+1，否则再与
    Figure PCTCN2022081190-appb-100032
    比较，以此类推；若最终未能计算出来向或去向车辆临时车道号TCL i的值，则设置默认车道号99，若车道号不等于99，执行步骤S03，否则判断车辆信息中的下一辆车是否异常变道；
    S03：查看目标车辆所在的路段是否为禁止变道路段，若是，执行步骤 S04，否则判断车辆信息中的下一辆车是否异常变道；
    S04：对计算出的目标车辆临时车道号TCL i，若TCL i为99，则不更新RCL i值；否则对比RCL i中存储的值，若RCL i中的值不等于99，且TCL i与RCL i的值不等，则临时计数器TSC i值加1，若RCL i中的值等于99，则使用TCL i更新RCL i的值；
    S05：比较临时计数器TSC i的值与预设阈值的大小，若大于阈值，判定该车辆存在异常变道行驶行为，更新异常变道发生的时间T i、目标车辆编号CN i、目标车辆车道号CL i，使用TCL i更新RCL i的值；否则判断车辆信息中的下一辆车是否异常变道。
  5. 根据权利要求3所述的一种基于毫米波雷达的车辆异常行驶行为识别方法，其特征在于，所述判断车辆逆行的方法包括：
    S11：对于来向车辆，计算车辆横向位置坐标
    Figure PCTCN2022081190-appb-100033
    与RY m之间的距离差，取绝对值后与
    Figure PCTCN2022081190-appb-100034
    比较大小，若小于
    Figure PCTCN2022081190-appb-100035
    则判定当前车辆车道号为k，否则再与
    Figure PCTCN2022081190-appb-100036
    比较大小，若小于
    Figure PCTCN2022081190-appb-100037
    则判定当前车辆车道号为k+1，否则再与
    Figure PCTCN2022081190-appb-100038
    比较，以此类推，若最终未能计算出目标车辆车道号的值，则设置默认车道号99；
    S12：计算车辆横向位置坐标
    Figure PCTCN2022081190-appb-100039
    与RY m之间的距离差，取绝对值后与
    Figure PCTCN2022081190-appb-100040
    比较大小，若小于
    Figure PCTCN2022081190-appb-100041
    则判定当前车辆车道号为k，否则再与
    Figure PCTCN2022081190-appb-100042
    比较大小，若小于
    Figure PCTCN2022081190-appb-100043
    则判定当前车辆车道号为k+1，否则再与
    Figure PCTCN2022081190-appb-100044
    比较，以此类推，若最终未能计算出目标车辆车道号的值，则设置默认车道号99；
    S13：对于车道号不等于99的目标车辆V i，若目标车辆V i属于右侧车辆，并且车辆纵向速度
    Figure PCTCN2022081190-appb-100045
    小于0，即右侧目标车辆存在逆向行驶可能性，临时计数器TSN i加1；若目标车辆V i属于左侧车辆，并且车辆纵向速度
    Figure PCTCN2022081190-appb-100046
    大于0，即右侧目标车辆存在逆向行驶可能性，临时计数器TSN i加1；
    S14：比较临时计数器TSN i的值与给定阈值的大小，若等于阈值，则判定为逆向行驶行为，更新目标车辆编号CN i、逆向行驶发生的时间T i、目标车辆车道号CL i，将目标车辆异常行驶行为发生后持续的时长NT i加1；若TSN i的值大于阈值，则只更新目标车辆异常行驶行为发生后持续的时长NT i，将NT i的值加1；若小于阈值，则判断车辆信息中的下一辆车是否逆行。
  6. 根据权利要求3所述的一种基于毫米波雷达的车辆异常行驶行为识别方法，其特征在于，所述判断车辆异常超速的方法包括：
    S21：对于来向车辆，比较车辆横向位置坐标
    Figure PCTCN2022081190-appb-100047
    与
    Figure PCTCN2022081190-appb-100048
    之间的大小，若
    Figure PCTCN2022081190-appb-100049
    小于
    Figure PCTCN2022081190-appb-100050
    判定车辆所在的路段号为m；否则再与
    Figure PCTCN2022081190-appb-100051
    比较大小，若小于
    Figure PCTCN2022081190-appb-100052
    判定车辆所在的路段号为m+1，否则再与
    Figure PCTCN2022081190-appb-100053
    比较，以此类推；对于去向车辆，比较车辆横向位置坐标
    Figure PCTCN2022081190-appb-100054
    与
    Figure PCTCN2022081190-appb-100055
    之间的大小，若
    Figure PCTCN2022081190-appb-100056
    小于
    Figure PCTCN2022081190-appb-100057
    判定车辆所在的路段号为m；否则再与
    Figure PCTCN2022081190-appb-100058
    比较大小，若小于
    Figure PCTCN2022081190-appb-100059
    判定车辆所在的路段号为m+1，否则再与
    Figure PCTCN2022081190-appb-100060
    比较，以此类推；若最终未能计算出来向或去向车辆所在路段号的值，则设置默认路段号99；
    S22：若目标车辆V i的的路段号不等于99，比较目标车辆V i的纵向速度
    Figure PCTCN2022081190-appb-100061
    与该车辆所在的路段的限速阈值，若小于限速阈值，则执行步骤S23，否则临时计数器TSO i加1；
    S23：对于来向车辆，计算车辆横向位置坐标
    Figure PCTCN2022081190-appb-100062
    与RY m之间的距离差，取绝对值后与
    Figure PCTCN2022081190-appb-100063
    比较大小，若小于
    Figure PCTCN2022081190-appb-100064
    则判定当前车辆车道号为k，否则再与
    Figure PCTCN2022081190-appb-100065
    比较大小，若小于
    Figure PCTCN2022081190-appb-100066
    则判定当前车辆车道号为k+1，否则再与
    Figure PCTCN2022081190-appb-100067
    比较，以此类推；对于去向车辆，计算车辆横向位置坐标
    Figure PCTCN2022081190-appb-100068
    与RY m之间的距离差，取绝对值后与
    Figure PCTCN2022081190-appb-100069
    比较大小，若小于
    Figure PCTCN2022081190-appb-100070
    则判定当前车辆车道 号为k，否则再与
    Figure PCTCN2022081190-appb-100071
    比较大小，若小于
    Figure PCTCN2022081190-appb-100072
    则判定当前车辆车道号为k+1，否则再与
    Figure PCTCN2022081190-appb-100073
    比较，以此类推；若最终未能计算出来向或去向车辆车道号的值，则设置默认车道号99；
    S24：比较临时计数器TSO i的值与给定阈值的大小，若小于阈值，则判断车辆信息中的下一辆车是否异常超速；若等于阈值，则判定存在异常超速行为，更新目标车辆编号CN i、异常超速发生的时间T i，目标车辆的车道号CL i，目标车辆异常行驶行为发生后持续的时长OT i加1；若TSO i的值大于阈值，则只更新目标车辆异常超速行为发生后持续的时长OT i，将OT i的值加1。
  7. 根据权利要求3所述的一种基于毫米波雷达的车辆异常行驶行为识别方法，其特征在于，所述判断车辆紧急制动的方法包括：
    S31：对于来向车辆，计算车辆横向位置坐标
    Figure PCTCN2022081190-appb-100074
    与RY m之间的距离差，取绝对值后与
    Figure PCTCN2022081190-appb-100075
    比较大小，若小于
    Figure PCTCN2022081190-appb-100076
    则判定当前车辆车道号为k，否则再与
    Figure PCTCN2022081190-appb-100077
    比较大小，若小于
    Figure PCTCN2022081190-appb-100078
    则判定当前车辆车道号为k+1，否则再与
    Figure PCTCN2022081190-appb-100079
    比较，以此类推；对于去向车辆，计算车辆横向位置坐标
    Figure PCTCN2022081190-appb-100080
    与RY m之间的距离差，取绝对值后与
    Figure PCTCN2022081190-appb-100081
    比较大小，若小于
    Figure PCTCN2022081190-appb-100082
    则判定当前车辆车道号为k，否则再与
    Figure PCTCN2022081190-appb-100083
    比较大小，若小于
    Figure PCTCN2022081190-appb-100084
    则判定当前车辆车道号为k+1，否则再与
    Figure PCTCN2022081190-appb-100085
    比较，以此类推；若最终未能计算出来向或去向车辆车道号的值，则设置默认车道号99；
    S32：设置临时计数器TSEB i值为0，记录目标车辆V i的速度
    Figure PCTCN2022081190-appb-100086
    至RD 2，临时计数器TSEB i的值加1；
    S33：若TSEB i等于间隔帧数FN，则将RD 1的值更新为RD 2的值，将当前目标车辆V i的速度
    Figure PCTCN2022081190-appb-100087
    记录至RD 2，再比较RD 1和RD 2的值，若满足RD 2大于RD 1乘以预定阈值，判断车辆信息中的下一辆车是否紧急制动， 否则判定当前目标车辆V i存在紧急制动行驶行为，更新目标车辆编号CN i、紧急制动发生的时间T i、目标车辆的车道号CL i，将TSEB i的值清0。
  8. 根据权利要求1所述的一种基于毫米波雷达的车辆异常行驶行为识别方法，其特征在于，所述车辆信息存储单元存储车辆的基本信息以及异常行驶信息；所述基本信息包括车辆实时位置P i、速度信息S i，车道信息RCL i；所述异常行驶信息包括异常变道行为C i、异常超速行为O i、逆行行为N i、紧急制动行为EB i，异常行驶行为发生的时间T i，异常超速行驶行为发生后持续的时长OT i，逆行行驶行为发生后持续的时长NT i，车辆编号CN i，车辆目标所处的车道号CL i，其中速度信息包括纵向速度
    Figure PCTCN2022081190-appb-100088
    和横向速度
    Figure PCTCN2022081190-appb-100089
    实时位置包括纵向位置坐标
    Figure PCTCN2022081190-appb-100090
    和横向位置坐标
    Figure PCTCN2022081190-appb-100091
    i表示实时检测跟踪到的车辆编号，车道信息RCL i初始值为99。
PCT/CN2022/081190 2021-12-14 2022-03-16 一种基于毫米波雷达的车辆异常行驶行为识别方法 WO2023108932A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202111525394.9A CN114495483A (zh) 2021-12-14 2021-12-14 一种基于毫米波雷达的车辆异常行驶行为识别方法
CN202111525394.9 2021-12-14

Publications (1)

Publication Number Publication Date
WO2023108932A1 true WO2023108932A1 (zh) 2023-06-22

Family

ID=81493578

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/081190 WO2023108932A1 (zh) 2021-12-14 2022-03-16 一种基于毫米波雷达的车辆异常行驶行为识别方法

Country Status (2)

Country Link
CN (1) CN114495483A (zh)
WO (1) WO2023108932A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115482672B (zh) * 2022-08-11 2023-12-19 西安天和防务技术股份有限公司 车辆逆行检测的方法、装置、终端设备及存储介质

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2007257536A (ja) * 2006-03-24 2007-10-04 Fujitsu Ltd ミリ波レーダによる道路交通監視装置
CN103680138A (zh) * 2012-09-25 2014-03-26 西安思丹德信息技术有限公司 一种车辆路况监测方法
CN111325978A (zh) * 2020-02-27 2020-06-23 江苏大学 一种高速公路车辆异常行为的全程监测警示系统和方法
CN112735137A (zh) * 2021-01-07 2021-04-30 奥谱毫芯(深圳)科技有限公司 基于毫米波信号的量化交通预警方法、装置、系统及介质
WO2021196144A1 (zh) * 2020-04-02 2021-10-07 华为技术有限公司 一种异常驾驶行为识别方法
CN113627213A (zh) * 2020-05-06 2021-11-09 浙江宇视科技有限公司 车辆异常行为监控方法、装置及系统

Family Cites Families (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7859392B2 (en) * 2006-05-22 2010-12-28 Iwi, Inc. System and method for monitoring and updating speed-by-street data
CN104424808B (zh) * 2013-08-27 2016-12-28 上海博泰悦臻电子设备制造有限公司 一种导航提示方法及装置、导航系统
CN103761229B (zh) * 2013-10-15 2017-11-07 北京航空航天大学 数据处理方法及装置
CN106571042A (zh) * 2015-10-12 2017-04-19 深圳市赛格导航科技股份有限公司 一种可变限速的车辆超速判定方法及系统
JP6953805B2 (ja) * 2017-06-01 2021-10-27 スズキ株式会社 路面情報収集システム
CN109410646A (zh) * 2018-12-22 2019-03-01 江苏宁沪高速公路股份有限公司 一种智能防撞预警系统
CN110807932A (zh) * 2019-11-12 2020-02-18 汤建强 一种车辆实线变道检测方法
CN112017447B (zh) * 2020-08-20 2022-09-23 北京赛博星通科技有限公司 一种基于gps位置信息判定车辆逆行违规的方法及系统
CN112415552B (zh) * 2020-11-17 2022-04-15 北京百度网讯科技有限公司 车辆位置的确定方法、装置及电子设备
CN112562329A (zh) * 2020-11-28 2021-03-26 山西省交通科技研发有限公司 一种基于毫米波雷达数据的交通事件感知方法
CN113191342A (zh) * 2021-07-01 2021-07-30 中移(上海)信息通信科技有限公司 车道定位方法及电子设备

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2007257536A (ja) * 2006-03-24 2007-10-04 Fujitsu Ltd ミリ波レーダによる道路交通監視装置
CN103680138A (zh) * 2012-09-25 2014-03-26 西安思丹德信息技术有限公司 一种车辆路况监测方法
CN111325978A (zh) * 2020-02-27 2020-06-23 江苏大学 一种高速公路车辆异常行为的全程监测警示系统和方法
WO2021196144A1 (zh) * 2020-04-02 2021-10-07 华为技术有限公司 一种异常驾驶行为识别方法
CN113627213A (zh) * 2020-05-06 2021-11-09 浙江宇视科技有限公司 车辆异常行为监控方法、装置及系统
CN112735137A (zh) * 2021-01-07 2021-04-30 奥谱毫芯(深圳)科技有限公司 基于毫米波信号的量化交通预警方法、装置、系统及介质

Also Published As

Publication number Publication date
CN114495483A (zh) 2022-05-13

Similar Documents

Publication Publication Date Title
CN110418745B (zh) 用于车辆护航的间隙测量
US9132837B2 (en) Method and device for estimating the number of lanes and/or the lane width on a roadway
US8577552B1 (en) Forward collision warning system with road-side target filtering
CN101916516B (zh) 一种车道偏移识别方法
CN110296713A (zh) 路侧自动驾驶车辆定位导航系统及单个、多个车辆定位导航方法
CN112859062B (zh) 一种基于雷达的车辆排队长度检测方法及系统
Homm et al. Fusion of laserscannner and video based lanemarking detection for robust lateral vehicle control and lane change maneuvers
JP2007066047A (ja) 道路形状認識装置
WO2020259301A1 (zh) 交叉口车辆延误时间与停车次数的计算方法和装置
CN105931467A (zh) 一种跟踪目标的方法及装置
CN112540365B (zh) 一种评估方法、装置、设备及存储介质
CN210119237U (zh) 路侧自动驾驶车辆定位导航系统
WO2023108932A1 (zh) 一种基于毫米波雷达的车辆异常行驶行为识别方法
CN105205805A (zh) 基于视觉的智能车辆横向控制方法
US20220314979A1 (en) Apparatus and Method for Controlling Driving of Vehicle
US11423661B2 (en) Object recognition apparatus
CN115856872A (zh) 一种车辆运动轨迹连续跟踪方法
CN116394892A (zh) 车辆的横向制动方法、装置、设备及存储介质
CN106405539A (zh) 车辆雷达系统以及用于去除一非感兴趣目标的方法
Polychronopoulos et al. Integrated object and road border tracking using 77 GHz automotive radars
CN115966084B (zh) 全息路口毫米波雷达数据处理方法、装置及计算机设备
CN110648542A (zh) 基于方位识别窄波雷达的高精度车流量检测系统
JP6851111B2 (ja) 踏切障害物検知装置
Vanpoperinghe et al. Model-based detection and tracking of vehicle using a scanning laser rangefinder: A particle filtering approach
CN117250595B (zh) 一种车载毫米波雷达金属井盖目标误报抑制方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22905703

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE