WO2024037375A1 - 交通区域锁控制方法、装置、设备和存储介质 - Google Patents

交通区域锁控制方法、装置、设备和存储介质 Download PDF

Info

Publication number
WO2024037375A1
WO2024037375A1 PCT/CN2023/111650 CN2023111650W WO2024037375A1 WO 2024037375 A1 WO2024037375 A1 WO 2024037375A1 CN 2023111650 W CN2023111650 W CN 2023111650W WO 2024037375 A1 WO2024037375 A1 WO 2024037375A1
Authority
WO
WIPO (PCT)
Prior art keywords
area
target
vehicle
target vehicle
lock
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.)
Ceased
Application number
PCT/CN2023/111650
Other languages
English (en)
French (fr)
Inventor
李湛
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
China Mobile Communications Group Co Ltd
China Mobile Group Hebei Co Ltd
Original Assignee
China Mobile Communications Group Co Ltd
China Mobile Group Hebei Co Ltd
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 China Mobile Communications Group Co Ltd, China Mobile Group Hebei Co Ltd filed Critical China Mobile Communications Group Co Ltd
Publication of WO2024037375A1 publication Critical patent/WO2024037375A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/16—Anti-collision systems
    • G08G1/167—Driving aids for lane monitoring, lane changing, e.g. blind spot detection
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/16—Anti-collision systems
    • 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

Definitions

  • the present disclosure belongs to the field of automatic driving technology, and in particular relates to a traffic area lock control method, device, equipment and storage medium.
  • the autonomous driving system based on 5G edge computing measures the distance of surrounding vehicles through the sensors equipped on the vehicle and directly determines whether it can change lanes or overtake. Traffic accidents often occur due to multiple vehicles occupying the same position in the same lane from multiple different lanes. The problem.
  • Embodiments of the present disclosure provide a traffic area lock control method, device, equipment and storage medium, which can avoid frequent traffic accidents caused by multiple vehicles seizing the same position in the same lane from multiple different lanes.
  • inventions of the present disclosure provide a traffic area lock control method.
  • the method includes:
  • the area lock in the target area is controlled to be in the occupied state, and the target vehicle is controlled to change lanes from the area where the target vehicle is located to the target area.
  • an embodiment of the present disclosure provides a traffic area lock control device, which includes:
  • the determination module is used to determine the area where the target vehicle is located based on the multi-dimensional spatio-temporal position
  • the determination module is also used to determine the offset area based on the projection area and the preset offset of the area where the target vehicle is located;
  • the determination module is also used to determine the target area based on the projection area and offset area;
  • a control module used to control the area lock in the target area to be occupied when the area where the target vehicle is located is in the adjacent area of the target area and the area lock of the target area is in an idle state, and control the target vehicle to move from the area where the target vehicle is located. Change lanes to target area.
  • a traffic area lock control device which includes: a processor and a memory storing computer program instructions;
  • the method described in the first aspect is implemented when the processor executes the computer program instructions.
  • embodiments of the present disclosure provide a computer storage medium.
  • Computer program instructions are stored on the computer-readable storage medium.
  • the method described in the first aspect is implemented.
  • embodiments of the present disclosure provide a computer program product, which is characterized in that, when instructions in the computer program product are executed by a processor of an electronic device, the electronic device causes the electronic device to execute the method described in the first aspect.
  • the traffic area lock control method, device, equipment and storage medium can obtain the multi-dimensional spatio-temporal position of the target vehicle; and then determine the area where the target vehicle is located based on the multi-dimensional spatio-temporal position; and then determine the area where the target vehicle is located based on the location of the target vehicle.
  • the projection area and the preset offset are used to determine the offset area; then the target area is determined based on the projection area and offset area; when the area where the target vehicle is located is adjacent to the target area, and the area lock of the target area is idle. , control the area lock in the target area to be occupied, and control the target vehicle to change lanes from the area where the target vehicle is located to the target area.
  • the target area of the target vehicle is determined, the area lock in the target area is controlled to change from the idle state to the occupied state, and the target vehicle is controlled to change lanes from the area where the target vehicle is located to the target area, so that the The target area is only for target vehicles to change lanes, which can avoid traffic accidents caused by multiple vehicles occupying the same position in the same lane from multiple different lanes.
  • Figure 1 shows a schematic diagram of controlling an autonomous vehicle to change lanes based on a mobile edge computing node provided by an embodiment of the present disclosure
  • Figure 2 shows a schematic flowchart of registering an autonomous vehicle as a user equipment to access a mobile edge computing node according to an embodiment of the present disclosure
  • Figure 3 shows a schematic diagram of the sensor equipped in the autonomous vehicle transmitting the driving data of the autonomous vehicle to the user plane functional module of the mobile edge computing node provided by the embodiment of the present disclosure
  • Figure 4 is a schematic flowchart of a traffic area lock control method provided by an embodiment of the present disclosure
  • Figure 5 shows a schematic diagram of acquiring a partial projection area based on any external sensor on the target vehicle provided by an embodiment of the present disclosure
  • Figure 6 shows a schematic diagram of planar projection based on multi-dimensional spatiotemporal positions provided by an embodiment of the present disclosure
  • Figure 7 shows a schematic diagram of a region lock request list provided by an embodiment of the present disclosure
  • Figure 8 shows a schematic flow chart of the training method of the optimal solution solution model provided by the embodiment of the present disclosure
  • Figure 9 is a schematic structural diagram of a traffic area lock control device provided by an embodiment of the present disclosure.
  • FIG. 10 is a schematic structural diagram of another traffic area lock control device provided by an embodiment of the present disclosure.
  • Figure 11 is a schematic structural diagram of a traffic area lock control device provided by an embodiment of the present disclosure.
  • the fifth generation mobile communication technology (5th Generation Mobile Communication Technology, 5G) is an innovation and change in network technology.
  • the 5G communication network is more decentralized. Small-scale or portable data centers are deployed at the edge of the network, and a part of the services requested by the terminal are handed over to local computing processing to meet the ultra-low latency requirements of ultra-reliable low-latency communication and massive large-scale connections to the Internet of Things. Latency requirements.
  • Mobile Edge Computing is one of the core technologies of 5G.
  • Mobile edge computing runs at the edge of the network and usually has high computing power, and is especially suitable for computing tasks that analyze and process large amounts of data. Because the mobile edge computing nodes are geographically very close to users and information sources, the network delay of MEC nodes in responding to user requests is greatly reduced, reducing the possibility of network congestion in the transmission network and core network, achieving high reliability and low cost Delayed transmission requirements.
  • the International Tecommunication Union-Radiocommunica has defined three typical application scenarios of 5G, namely enhanced mobile broadband (eMBB) and ultra-reliable low-latency communication (Ultra Reliable). Low Latency Communication (uRLLC) and Massive Machine Type of Communication (mMTC).
  • eMBB enhanced mobile broadband
  • Ultra Reliable ultra-reliable low-latency communication
  • uRLLC Low Latency Communication
  • mMTC Massive Machine Type of Communication
  • eMBB is mainly targeted at high-bandwidth-demanding services such as Virtual Reality (VR), Augmented Reality (AR) and online high-definition video (4K/8K);
  • uRLLC is mainly targeted at vertical industries such as smart transportation and smart medical care.
  • mMTC is mainly targeted at IoT services with high connection density requirements for over 100 billion devices such as smart cities, smart industries and smart water meters.
  • the field of autonomous driving in smart transportation mainly involves autonomous vehicles and mobile edge computing working together to realize intelligent and comprehensive perception, judgment, reasoning and decision-making in various complex traffic environments.
  • the autonomous driving system mounted on an autonomous vehicle obtains its own status and surrounding traffic environment information through internal and external sensors equipped in the vehicle.
  • internal sensors mainly include Including vehicle speed sensors, acceleration sensors, wheel speed sensors and yaw rate sensors, etc.
  • External sensors include cameras, lidar, millimeter wave radar and positioning systems. These sensors collect all-round traffic environment information of self-driving vehicles and send the all-round traffic environment information of self-driving vehicles to the MEC node, which performs relevant traffic dispatch calculations. Subsequently, the MEC node returns the calculation results to the automatic Driving system, the autonomous driving system executes decision-making commands and controls the normal driving of autonomous vehicles, achieving the effect of ensuring smooth and orderly traffic while avoiding traffic jams or accidents. Similar to how a human driver perceives the driving environment and vehicle status through sensory systems such as vision, hearing, and touch during driving.
  • MEC nodes are mainly used to collect traffic environment information around autonomous vehicles and feed this traffic environment information back to the autonomous driving system of the autonomous vehicle for auxiliary decision-making.
  • this traffic environment information There is no reasonable and comprehensive use of this traffic environment information to carry out In-depth data mining analysis truly functions as an intelligent computing dispatch center.
  • the autonomous driving system of each vehicle makes local and independent judgments without considering the overall situation, which can easily cause multiple vehicles to compete for the same position in the same lane and cause traffic accidents.
  • the main technical means to avoid traffic accidents is to use sensors to measure the distance of surrounding vehicles and directly determine whether it is possible to change lanes or overtake. It is mainly suitable for handling simple and low-density traffic scenarios with single and double lanes, relatively idle and slow speeds. However, it is unable to handle complex and high-density traffic scenarios with many lanes, congestion and high speeds. It cannot avoid traffic accidents caused by multiple vehicles occupying the same position in the same lane from multiple different lanes. It cannot meet the needs of smart transportation. actual needs.
  • embodiments of the present disclosure provide a traffic area lock control method, device, equipment and storage medium. Before describing the traffic area lock control method provided by the embodiment of the present disclosure, the process of controlling a vehicle to change lanes based on the MEC node provided by the embodiment of the present disclosure is briefly introduced.
  • the cloud computing center node establishes communication with multiple MEC nodes that function as traffic area lock control and dispatch centers.
  • the MEC nodes establish communication with the autonomous driving system of each autonomous vehicle in multiple lanes.
  • the MEC node controls autonomous vehicle A and autonomous vehicle B.
  • Any vehicle changes lanes from the lane where the vehicle is located to the target area R1 of the target lane, and/or any autonomous vehicle among the autonomous vehicle C and the autonomous vehicle D changes lanes from the lane where the vehicle is located to the target area R2 of the target lane. .
  • the autonomous vehicle In order to control the autonomous vehicle to change lanes based on the MEC node, the autonomous vehicle needs to be registered as a user device and connected to the MEC node. As shown in Figure 2, before controlling an autonomous vehicle to change lanes based on an MEC node, the autonomous vehicle needs to be registered as a User Equipment (UE) within the 5G network signal coverage and connected to the MEC node.
  • UE User Equipment
  • the specific process of registering an autonomous vehicle as a UE to access the MEC node is as follows:
  • the UE initiates a registration request, and the Radio Access Network (RAN) where the MEC node is located receives the registration request and forwards the registration request to the Access and Mobility Management Function (AMF).
  • RAN Radio Access Network
  • AMF Access and Mobility Management Function
  • AMF selects forwarding based on local configuration or network slicing.
  • AMF relocation will be initiated; AMF receives the registration request and finds the corresponding authentication server function module (Subscription Concealed Identifier, SUCI) based on the routing code. Authentication Server Function, AUSF) makes an authentication request.
  • AMF Subscribed Identifier
  • AUSF Authentication Server Function
  • AUSF returns authentication parameters, which include random data Random Number, authentication data Authentication Token, and expected response data.
  • AMF sends the random data and authentication data in the authentication parameters to the UE; the UE uses the authentication data to authenticate the network, and then returns the response data to the AMF; the AMF compares the response data, and if it matches, the authentication is successful, and then The response data is sent to AUSF; AUSF compares the response data, and if it matches, returns the user permanent identifier (Subscription Perpetual Identifier, SUPI) corresponding to SUCI to AMF.
  • SUPI Subscribescription Perpetual Identifier
  • AMF initiates registration to the Unified Data Management module (Unified Data Management, UDM), and subscribes to contract data management in UDM.
  • Unified Data Management UDM
  • the AMF needs to register with the UDM for terminal connection management again.
  • AMF uses the International Mobile Subscriber Identity Code (International Mobile Subscriber Identity Code) Identification Number, IMSI) parameter is used to initiate a request for subscription data to UDM, and UDM returns the IMSI subscription data.
  • IMSI International Mobile Subscriber Identity Code
  • AMF initiates a subscription to the terminal subscription data change to UDM based on the IMSI, and UDM returns a notification of successful subscription.
  • AMF will select a policy control function (Policy Control Function, PCF) based on SUPI.
  • Policy Control Function Policy Control Function
  • AMF can also adjust the association policy to PCF according to mobility restrictions.
  • the AMF returns a registration acceptance message to the UE.
  • the UE responds to complete registration and access to the 5G network MEC.
  • the sensors equipped in the autonomous vehicle need to transmit the driving data of the autonomous vehicle to the user plane function module (User Plane Function, UPF) in the MEC node.
  • UPF User Plane Function
  • the internal and external sensors equipped in the autonomous driving vehicle obtain its own status and surrounding environment information.
  • the sensors then transmit the collected comprehensive traffic environment information of the vehicle to the UPF in the MEC node through the autonomous driving system.
  • UPF is responsible for data transmission and data processing.
  • FIG. 4 is a schematic flowchart of a traffic area lock control method provided by an embodiment of the present disclosure. As shown in Figure 4, the traffic area lock control method provided by the embodiment of the present disclosure includes the following steps:
  • the target vehicle is controlled by the MEC node and will change lanes from the lane where the target vehicle is located to the lane where the area lock is located based on the status of the area lock.
  • Area lock is a traffic area lock that locks the target area.
  • the multi-dimensional spatio-temporal position of the target vehicle is determined based on the multi-dimensional spatio-temporal data of the target vehicle.
  • Multidimensional spatiotemporal data includes but is not limited to the current moment when the target vehicle is traveling and the latitude, longitude, altitude, traveling direction of the target vehicle at that moment, and the number of vehicles in the geographical area within the unit area.
  • the automatic driving vehicle A is used as the target vehicle for explanation.
  • the MEC node through the autonomous driving system, acquires the multi-dimensional spatio-temporal data of the target vehicle collected by the internal and external sensors equipped in the target vehicle, so as to determine The multi-dimensional space-time position of the target vehicle is determined, and then the area where the target vehicle is located is determined based on the multi-dimensional space-time position of the target vehicle.
  • S402. Determine the area where the target vehicle is located based on the multi-dimensional space-time position.
  • the area where the target vehicle is located includes the area where the target vehicle occupies the lane where the target vehicle is located.
  • the area where the target vehicle is located includes multi-dimensional spatio-temporal location. Determine the area where the target vehicle is located, so that the target area of the target vehicle can be determined at the MEC node based on the area where the target vehicle is located.
  • the area where the target vehicle occupies the lane where the target vehicle is located can be arbitrarily defined based on the multi-dimensional spatio-temporal position as the area where the target vehicle is located.
  • the area in which the target vehicle occupies the lane where the target vehicle is located can be defined as the area where the target vehicle is located.
  • step S402 in order to further refine the area where the target vehicle is located, so as to improve the efficiency of controlling the target vehicle to change lanes from the area where the target vehicle is located to the target area, step S402 may also include steps S4021 to step S4022.
  • S4021 According to the multi-dimensional space-time position, obtain the projection points of the target vehicle in multiple preset directions.
  • the preset directions include the left front, left rear, right front and right rear of the target vehicle.
  • At least one external sensor is equipped in any direction of the target vehicle's left front, left rear, right front, and right rear. Through the external sensors of the target vehicle, the data of at least four projection points on the left front, left rear, right front and right rear of the target vehicle are obtained, and then based on the data of the projection points, the plane where the target vehicle occupies the lane where the target vehicle is located is calculated. Projection area, use this plane projection area as the area where the target vehicle is located.
  • the external sensor may be a laser radar or millimeter wave radar type sensor, and the external sensor can emit laser, millimeter wave and other electromagnetic beams at all angles.
  • Induction strips are installed on both sides of each lane. The induction strips can accurately sense and receive each projection point of the electromagnetic wave beam emitted by the target vehicle.
  • the plane projection area occupied by the target vehicle in the lane where the target vehicle is located is calculated, and the plane projection area is used as the area where the target vehicle is located, achieving further accuracy of the area where the target vehicle is located. In order to improve the efficiency of controlling the target vehicle to change lanes from the area where the target vehicle is located to the target area.
  • S403. Determine the offset area according to the projection area and the preset offset of the area where the target vehicle is located. area.
  • the projection area of the area where the target vehicle is located is a positional relationship based on projection.
  • the target vehicle occupies the plane projection area of the lane where the target vehicle is located, and projects to the projection area of the lane where the target vehicle is about to change lanes.
  • the preset offset is the preset offset of the projection area of the area where the target vehicle is located when the target vehicle changes lanes to the planned lane change area.
  • the projection area of the area where the target vehicle is located is the plane projection of the multi-dimensional space-time position of the target vehicle. Based on the projection area and the preset offset, the offset area is calculated. Determine the offset area so that when the target vehicle changes lanes, it will accompany the driving process of the target vehicle. A safe distance between the target vehicle and the vehicles in front and behind the target vehicle is reserved in the target area of the target vehicle to avoid collisions between the target vehicle and the target vehicle. A rear-end collision occurred between the front and rear vehicles.
  • the preset offset may include a front offset and a rear offset of a planar projection of the multi-dimensional spatiotemporal position of the target vehicle.
  • the offset area includes a front offset area and a rear offset area. The front offset area is determined based on the projected area and the forward offset of the area, and the rear offset area is determined based on the projected area and the rear offset of the area.
  • the projection area of the area can also be obtained through positioning methods such as grid division or GPS.
  • positioning methods such as grid division or GPS.
  • problems such as inaccuracy due to too large grids, poor GPS signals, or large errors due to long distances may occur.
  • step S403 may also include steps S4031 to S4032.
  • S4031 Determine the projection area according to the positional relationship of the projection of the area where the target vehicle is located.
  • S4032 Determine the offset area based on the preset offset amount and projection area.
  • the projection area of the area where the target vehicle is located includes using an external sensor of the target vehicle as a projection point, sweeping at an angle of at least 90 degrees, emitting electromagnetic beams such as lasers or millimeter waves, and targeting sensors on both sides of the planned lane change area.
  • the strip is projected flatly to form a projection area.
  • planar projection is performed to the induction zones on both sides of the planned lane change area to form a projection area of the area where the target vehicle is located.
  • the above method of obtaining the projection area of the target vehicle through the positional relationship of the projection of the area where the target vehicle is located is based on the principle that electromagnetic waves with shorter wavelengths such as lasers or millimeter waves propagate close to linear propagation. Compared with positioning methods such as gridding or GPS, , can obtain the target more quickly, intuitively and accurately The projection area of the target vehicle avoids the problems of inaccurate grids, poor GPS signals, or large errors due to long distances.
  • the projection area is offset and expanded forward, and the expanded area is the forward offset area.
  • the projection area is offset to the rear according to the expansion of the rear offset amount in the preset offset amount, and the area with expanded offset is the rear offset area.
  • the projection area of the target vehicle is obtained, and the projection area is offset and expanded to determine the offset area, which can further accurately offset the area, so as to improve the control of the target vehicle to change lanes from the area where the target vehicle is located to efficiency in the target area.
  • Figure 6 shows a schematic diagram of planar projection based on multi-dimensional spatiotemporal positions provided by an embodiment of the present disclosure.
  • the target area includes a projection area and an offset area. Plane projection is performed based on the multi-dimensional spatiotemporal position.
  • the target area can be determined based on the projection area and offset area.
  • the relationship between the multidimensional spatiotemporal position, the plane projection of the multidimensional spatiotemporal position, the area where the target vehicle is located, and the target area is expressed as follows:
  • R' represents the area where the target vehicle is located;
  • R represents the target area of the target vehicle;
  • t i represents time;
  • w i represents the latitude of the target vehicle;
  • l i represents the longitude of the target vehicle;
  • h i represents the height of the target vehicle;
  • d i represents the driving direction of the target vehicle;
  • c i represents the number of vehicles in the geographical area within the unit area where the target vehicle is located.
  • the target area is represented as follows:
  • R represents the target area of the target vehicle
  • ⁇ represents the weighting parameter, which can be adjusted to expand the target area R in multiples
  • t i time
  • w i represents the latitude of the target vehicle within the range of time i
  • l i represents the longitude of the target vehicle within the range of time i
  • h i represents the time Within the range of i, the height of the target vehicle
  • d i represents the driving direction of the target vehicle within the range of time i
  • c i represents the number of vehicles in the geographical area per unit area where the target vehicle is located within the range of time i
  • ⁇ represents the weighted parameter, which can adjust the front
  • the forward and rear offsets increase as the speed of the target vehicle increases. For example, when the vehicle speed is lower than 20km/h, the offset remains 10 meters; when the vehicle speed is lower than 40km/h, the offset remains 40 meters; when the vehicle speed is lower than 60km/h, the offset remains 60 meters; when the vehicle speed is lower than 80km/h , the offset remains 80 meters; when the vehicle speed is lower than 100km/h, the offset remains 150 meters to 200 meters.
  • the front and rear offset areas are the product of the target vehicle's front and rear offset (i.e. distance) and the road width.
  • the front and rear offset of the target vehicle are at least the maximum allowable speed of the lane where the target vehicle is located and the product of time.
  • t i represents time
  • G F represents the forward offset of the target vehicle within the range of time i
  • V max represents the maximum allowable speed of the lane where the target vehicle is located.
  • the area where the target vehicle is located and the target area are dynamically corresponding areas within a time range determined based on the multi-dimensional spatiotemporal position of the target vehicle.
  • the area lock of the target area is in the idle state, which means that there is no area lock in the target area; the area lock in the target area is in the occupied state, which means that the area lock exists in the target area.
  • Controlling the area lock in the target area to be in the occupied state means that the MEC node adds the area lock of the target vehicle to the target area, so that the area lock of the target vehicle exists in the target area.
  • the projection area of the area where the target vehicle is located is within the range of the target area.
  • the MEC node adds the area lock of the target vehicle to the target area, so that the area of the target vehicle exists in the target area. lock, and control the target vehicle to change lanes from the area where the target vehicle is located to the target area.
  • the target vehicle By obtaining the multi-dimensional space-time position of the target vehicle; and then determining the target vehicle based on the multi-dimensional space-time position.
  • the target vehicle determines the target area of the target vehicle, control the area lock in the target area to change from the idle state to the occupied state, and control the target vehicle to change lanes from the area where the target vehicle is located to the target area, so that the target area is only for the target vehicle to change lanes, It can avoid traffic accidents caused by multiple vehicles occupying the same position in the same lane from multiple different lanes.
  • steps S406 to S407 may also be included.
  • S406 Obtain the vehicle density and traffic accident probability in the geographical area within the unit area where the target vehicle is located.
  • a multi-dimensional space-time vehicle density model is established.
  • the multidimensional space-time vehicle density model is expressed as follows:
  • a traffic accident early warning model is established based on the number of vehicles that have suffered traffic accidents in the geographical area within the unit area where the target vehicle is located and the number of all vehicles passing through the geographical area, that is, the historical statistical data of accidents in the traffic area.
  • the traffic accident early warning model is expressed as follows:
  • the number of vehicles in the area where traffic accidents occurred; c i represents the number of vehicles in the geographical area within the unit area where the target vehicle is located within the time i range.
  • the vehicle density in the geographical area within the unit area where the target vehicle is located is obtained.
  • the traffic accident early warning model the probability of traffic accidents in the geographical area within the unit area where the target vehicle is located is obtained.
  • the unit area can be preset and adjusted according to actual needs.
  • the vehicle density exceeds the vehicle density threshold, or the traffic accident probability exceeds the first traffic accident probability threshold, or the vehicle density exceeds the vehicle density threshold and the traffic accident probability exceeds the first traffic accident probability threshold it represents the target
  • the geographical area within the unit area where the vehicle is located is in a complex and high-density traffic scenario with many lanes, congestion and high speed.
  • M(c i ) represents the vehicle density in the geographical area within the unit area where the target vehicle is located; M ⁇ represents the initial threshold of vehicle density; P(h i ) represents the vehicle density within the unit area where the target vehicle is located within the range of time i.
  • the MEC node when the vehicle density exceeds the vehicle density threshold, or the traffic accident probability exceeds the first traffic accident probability threshold, or the vehicle density exceeds the vehicle density threshold and the traffic accident probability exceeds the first traffic accident probability
  • the MEC node can be automatically triggered to start, so that the MEC node can monitor whether the area where the target vehicle is located is in the adjacent area of the target area, and the area lock of the target area is idle, and the area lock in the target area is controlled to be occupied by the target vehicle.
  • the occupied state of the vehicle is occupied, and the target vehicle is controlled to change lanes from the area where the target vehicle is located to the target area.
  • the MEC node startup may not be triggered, allowing users to flexibly control the movement of the target vehicle.
  • the vehicle density and traffic accident probability calculated by formula (4) and formula (5) can be combined to optimize the target area range that needs to be locked. If formula (6) is reached conditions, by adjusting the weighting parameters in formula (2), appropriately expand the scope of the target area, that is, the scope of the area lock, thereby reducing vehicle density and the probability of traffic accidents.
  • the vehicle density exceeds the vehicle density threshold, or the traffic accident probability exceeds the first traffic accident probability threshold, or the vehicle density exceeds the vehicle density threshold and the traffic accident probability exceeds the first traffic accident probability threshold it is automatically triggered.
  • the MEC node When the MEC node is activated, it can accurately handle the problem of traffic accidents caused by multiple vehicles occupying the same position in the same lane from multiple different lanes in complex and high-density traffic scenarios with many lanes, congestion, and fast speeds.
  • the traffic area lock control method may also include steps S501 to 503.
  • the area lock request is an area lock request sent by the target vehicle based on the vehicle information of the target vehicle and the target area.
  • S503 When it is detected that the area lock of the target area is in an idle state and the area lock request of the target vehicle is reached according to the order in the area lock request list, control the target vehicle to change lanes from the area where the target vehicle is located to the target area.
  • the target vehicle needs to first send the target vehicle to the MEC node based on the vehicle information of the target vehicle and the area lock request sent by the target area. For example, if the area lock request includes the license plate number XXX, please add area lock K in the target area R of the target lane to be occupied.
  • the MEC node Based on the area locks of other vehicles in the target area, the MEC node will manage the area lock requested for the target area through the queue mode, and add the area lock request to the area lock request list of the target area (i.e., area lock request queue) middle.
  • the MEC node directly allows the target vehicle to add an area lock to the target area.
  • the MEC node directly allows the target area
  • the target vehicle adds area lock K i to the target area.
  • the MEC node adds the area lock request to the area lock request list of the target area according to the time sequence of receiving the area lock request. middle.
  • the MEC node When it is detected that the area lock of the target area is in an idle state, and the area lock request of the target vehicle is reached according to the order in the area lock request list, that is, the MEC node detects that there are no area locks of other vehicles in the target area, and in the area of the target vehicle After the area lock before the lock fails or is released, the MEC node controls the target vehicle to change lanes from the area where the target vehicle is located to the target area.
  • the MEC node controls the locking process of the area lock of the target vehicle.
  • the target vehicle plans to occupy the designated traffic area (i.e., the target area)
  • it sends an area lock request to the MEC node and applies for locking the target area.
  • the MEC node monitors whether the area lock of the target area is in an idle state. If the area lock of the target area is in an idle state, it updates the area lock of the target area to the occupied state occupied by the target vehicle. Subsequently, the MEC node returns a successful response result to the target vehicle and controls the target vehicle to drive into the target area.
  • the MEC node adds the area lock request to the area lock request list of the target area.
  • the region lock request list contains the maximum length of the queue. After the MEC node adds the area lock request to the area lock request list of the target area, if the current area lock request list does not exceed the maximum length of the queue, the MEC node returns to the queue waiting state to the target vehicle; if the current area lock request list exceeds If the maximum length of the queue is exceeded, the MEC node returns the failure result and the reason to the automatic driving system of the target vehicle.
  • the MEC node controls the unlocking process of the target vehicle's area lock. Unlocking area locks is divided into active unlocking mode and passive unlocking mode.
  • the target vehicle actively applies for a zone lock unlock request to the zone lock control system.
  • the zone lock control system releases the zone lock for the target zone and updates the zone lock flag of the target zone to idle state.
  • the active unlocking mode mainly relies on the target vehicle to promptly and accurately apply to the area lock control system for unlocking based on its current location, which requires the target vehicle's Internal and external sensors, autonomous driving systems, and the 5G network they are located in have high sensitivity and performance.
  • the area lock control system In the passive unlocking mode, after the target vehicle leaves the target area, the area lock control system automatically determines the location of the target vehicle through 5G NR centimeter wave or millimeter wave positioning technology. When it is determined that the target vehicle has left the target area, the area lock control system releases the lock. The specified area lock, and updates the area lock flag of the target area to the idle state. The target vehicle passively accepts the processing results of the area lock control system.
  • the passive unlocking mode can effectively make up for the lack of active unlocking mode capabilities of the target vehicle. That is, when the target vehicle's internal and external sensors, autonomous driving system, and 5G network do not have high sensitivity and performance, the area lock control system will Computing pressure is mainly concentrated on the mobile edge computing side.
  • either active unlocking mode or passive unlocking mode can be used, or a combination of each other can be used to achieve rapid area unlocking.
  • a target vehicle under normal circumstances, is set to hold at most one area lock by default.
  • the size of the lock granularity that is, the area of the target area locked by the area lock, can be the minimum plane area of the target vehicle.
  • a target vehicle can also apply for multiple area locks.
  • the setting of the number of area locks of the target vehicle can be controlled.
  • the area lock request in order to prioritize rescue vehicles such as 110 police cars, 120 medical ambulances, and 119 fire trucks for lane changes, the area lock request also includes the vehicle type of the target vehicle.
  • the traffic area provided by embodiments of the present disclosure
  • the lock control method may also include steps S601 to S602.
  • S601 When receiving the area lock request of the target vehicle, determine the priority of the target vehicle based on the preset correspondence between vehicle type and priority.
  • S602 Add the region lock request to the region lock request list based on priority.
  • weighting the area lock By weighting the area lock, different weights are given to the area lock of the target vehicle in emergency situations, and different priorities are assigned to the target vehicle according to the different weights of the area lock. That is, different preset vehicle types are assigned different preset Vehicle types target vehicles with different priorities. The greater the area lock weight of the target vehicle and the more urgent the traffic situation corresponding to the target vehicle, the higher the priority of the target vehicle.
  • Preset vehicle types include priority vehicles and regular vehicles.
  • Priority vehicles include but are not limited to 110 police cars, 120 medical ambulances, 119 fire trucks and other rescue vehicles. Priority vehicles can be further subdivided according to the different rescue attributes of the vehicle. For example, from high to low, the priorities of 120 medical ambulance, 119 fire truck and 110 police car are set to level one, level two and level three. When the 120 medical ambulance, 119 fire truck and 110 police car send area lock requests at the same time, Priority will be given to allowing the 120 medical ambulance to change lanes. After the 120 medical ambulance successfully changes lanes, the 119 fire truck and 110 police car will be allowed in sequence.
  • the area lock request corresponding to the vehicle with the highest priority is added to the frontmost position in the area request list, until the MEC node detects that there is no area for other vehicles in the target area. lock, and after the area lock before the area lock of the vehicle with the highest priority level becomes invalid or released, the MEC node controls the vehicle with the highest priority level to change lanes from the area where the vehicle is located to the target area.
  • self-driving vehicle C is a regular vehicle
  • self-driving vehicle D is a priority vehicle.
  • self-driving vehicle C and self-driving vehicle D change lanes from the vehicle lane to the target area R2 of the target lane
  • self-driving vehicle D has a higher priority than self-driving vehicle C
  • self-driving vehicle D has a higher priority than self-driving vehicle C
  • the MEC node when the area lock of the target area is in the occupied state, when the MEC node receives the area lock request sent by the autonomous vehicle C and the autonomous vehicle D, since the priority of the autonomous vehicle D is higher than that of the autonomous vehicle C.
  • the area lock request list self-driving vehicle D is arranged in front of self-driving vehicle C.
  • the MEC node assigns different priority area locks to different target vehicles according to the preset vehicle type, and prioritizes priority vehicles for lane changes. , which can ensure that rescue vehicles can quickly pass through this road section, reduce waiting time in queues, and avoid missing the best rescue time due to waiting.
  • area locks of all vehicles are assigned different weights, there may be a problem that target vehicles with smaller area lock weights are always waiting in line. Therefore, area locks for conventional vehicles other than priority vehicles are not prioritized, but are scheduled and processed by the MEC node on a first-request-first-serve basis.
  • First request first served queue processing logic For area lock requests for the same target area in the same target lane initiated by conventional vehicles, use First request first served queue processing logic.
  • the target vehicle leaves the target area of the target lane, the area lock at the head of the corresponding queue is released, and the area lock of the next target vehicle occupies the target area. Then the next target vehicle can choose to drive into the target area of the target lane. . If the next target vehicle gives up driving into the target area of the target lane, the area lock of the target vehicle will be invalid, and the area lock of the following target vehicle will occupy the target area in turn, and a similar process as above will be performed.
  • the area lock is exclusive, that is, before the area lock is released or becomes invalid, other vehicles are not allowed to add area locks to the target area of the target lane and need to wait in line.
  • the traffic area lock control method provided by the embodiment of the present disclosure may also include the steps S603 to step S606.
  • Step S603 Obtain the first request time, which is the request time when the region lock request is added to the region lock request list.
  • Step S604 Obtain the second request time, which is the request time when the area lock request of the vehicle currently occupying the target area is added to the area lock request list.
  • Step S605 Calculate the difference between the first request time and the second request time to obtain the waiting time of the target vehicle.
  • Step S606 Control the target vehicle to change lanes from the area where the target vehicle is located to the target area after waiting for a period of time.
  • the waiting time is the request time for the area lock request to be added to the area lock request list, and the length of time from the request time for the area lock request of the vehicle currently occupying the target area to be added to the area lock request list.
  • the waiting time can be calculated. However, during actual driving, the waiting time is still affected by multiple factors such as the location of the target area, the size of the target area, the current vehicle speed, and road traffic conditions.
  • weighted parameters ⁇ , ⁇ and ⁇ , thresholds M ⁇ and P ⁇ , the number of vehicles c i and the waiting time T w involved in the above traffic area lock control method can all be adjusted through continuous iterative calculations of the machine learning algorithm. , try to meet the requirements of safely accommodating more vehicles within the geographical area of the unit area where the target vehicle is located, the target area locking range being small and accurate, and the waiting time being shortened to avoid traffic congestion.
  • the traffic area lock control method provided by the embodiment of the present disclosure may also include step S701 to step S702.
  • S701 Input the traffic accident probability value, vehicle density and target area into the optimal solution solution model to obtain the weighted parameters corresponding to the adjusted target area;
  • S702 Adjust the target area according to the weighted parameters corresponding to the adjusted target area, and obtain the adjusted target area.
  • the traffic area lock control method it is required that the probability of traffic accidents in autonomous driving is low, the vehicle density is high, the locked target area is small, and the waiting time is short.
  • the above problems are reduced to multi-objective optimization problems through data modeling, and a multi-objective function optimization model is established.
  • the traffic accident probability value, vehicle density and target area are input into the optimal solution solution model, and the optimal solution solution model is input to real-time and dynamically adjust the values of the weighted parameters ⁇ , ⁇ and ⁇ corresponding to the target area, and / Or the values of the thresholds M ⁇ and P ⁇ and the number of vehicles c i , and iterative calculations are performed to obtain the weighted parameters corresponding to the adjusted target area that satisfies the multi-objective function optimization model.
  • the MEC node when the MEC node predicts a relatively congested traffic situation, the MEC node solves the model through the optimal solution, uses the multi-objective function optimization model as the stopping condition of the iterative calculation, and obtains the optimal weighted parameters ⁇ and ⁇ through iterative calculation. and ⁇ , the optimal values of thresholds M ⁇ and P ⁇ and the number of vehicles c i to regulate traffic until the traffic returns to normal in the future. Then, combine the calculation The optimal value of is calculated to obtain the traffic area locking range, that is, the target area.
  • the efficiency of the target vehicle changing lanes from the area where the target vehicle is located to the target area can be improved.
  • step S405 may also include step S4051.
  • Lane changing based on the adjusted target area can achieve the goals of safely accommodating more vehicles in the geographical area within the unit area of the target vehicle, having a small and accurate target area locking range, and shortening the waiting time, further improving the target vehicle's efficiency.
  • the training method of the optimal solution model includes steps S801 to S805.
  • S801 Obtain the traffic accident probability values, vehicle density and initialization parameters of the target area corresponding to the target vehicles in all lanes.
  • S803 Based on the Sparrow search algorithm, update the initialization parameter matrix according to the second traffic accident probability threshold.
  • the multi-objective function optimization model is expressed as follows:
  • f. represents multiple objective functions in the multi-objective function optimization model; Indicates the target area with the smallest and most accurate locking range; Indicates the minimum waiting time; Indicates the lowest probability of traffic accident; represents the highest vehicle density; st represents multiple constraints in the multi-objective function optimization model.
  • the optimal solution solution model may be a convolutional neural network algorithm improved based on the Sparrow search algorithm.
  • the optimal solution solution model can also be based on other artificial intelligence algorithms and machine learning algorithms, and is not limited here.
  • the sparrow population is divided into discoverers and joiners.
  • the discoverers are responsible for finding food in the population and providing foraging areas and directions for the entire sparrow population, while the joiners use Finders come to obtain food, and sparrow populations will engage in anti-predator behavior when they sense danger.
  • the convolutional neural network is a feedforward neural network that uses convolution calculations and has a deep structure. It can process complex spatiotemporal sequence data, and iteratively guides the next hidden layer input and conducts the feedthrough results. Outcome prediction.
  • the improved convolutional neural network algorithm based on Sparrow search combines the advantages of Sparrow search algorithm and convolutional neural network algorithm. It can use the target vehicle of historical data as the Sparrow finder and provide guidance for subsequent target vehicles as joiners. If encounter Dangerous scenes with a high probability of traffic accidents are avoided by anti-predation behavior.
  • the convolutional neural network performs deep learning through the Sparrow search algorithm, and outputs the optimal value through iterative operations of the convolution layer, pooling layer and fully connected layer.
  • the convolutional neural network algorithm based on the improved Sparrow search algorithm will be used to find the optimal solution to the multi-objective function optimization model shown in formula (8), and the training process of the optimal solution solution model will be described. bright.
  • the position of the explorer (i.e. the discoverer) in the matrix Y is iteratively updated as follows:
  • a + represents a matrix with a random amplitude of 1 or -1 for each element.
  • Z l and Z l+1 represent the convolution input and output of the l+1 layer, that is, the feature map; K is the number of channels of the feature map, and f, s 0 and b represent the convolution layer parameters and deviation.
  • the output feature map is passed to the pooling layer for feature selection and information filtering.
  • the pooling layer model replaces the results of a single point in the feature map with the feature map statistics of its adjacent areas, Calculated as follows:
  • ⁇ means that the pre-specified parameter takes the mean or maximum value in the pooling area.
  • the fully connected layer performs a nonlinear combination on the features extracted by the convolutional layer to obtain the output.
  • the final output layer output includes values such as ⁇ , ⁇ , ⁇ , c i , etc.
  • the final output layer outputs values including ⁇ , ⁇ , ⁇ and c i , it corresponds to the goals of low probability of traffic accidents in autonomous driving, high vehicle density, small locked target area and short waiting time, that is, if the final output If the values output by the layer, including ⁇ , ⁇ , ⁇ , and c i, satisfy formula (8), then the values output by the last output layer, including ⁇ , ⁇ , ⁇ , and c i , are accepted as the optimal values. Otherwise, continue to improve based on sparrow search.
  • the convolutional neural network algorithm is used for iterative solution.
  • the output includes ⁇ , ⁇ , ⁇ and c i and other values that satisfy the multi-objective function optimization model, reaching the target area where the probability of traffic accidents in autonomous driving is low, the vehicle density is high, and the target area is locked. Targets that are small and have short waiting times can further improve the efficiency of the target vehicle changing lanes from the area where the target vehicle is located to the target area.
  • the traffic area lock control method provided by the embodiment of the present disclosure may further include step S901.
  • S901 Obtain and send the lane change failure rate of the target vehicle.
  • the MEC node obtains the lane change failure rate of the target vehicle based on historical data, and automatically transmits the lane change failure rate to the road transportation management department, which is then handed over to the road transportation management department for big data analysis so that it can be considered by the road transportation management department.
  • appropriate improvement measures such as proper diversion of traffic if the road is overly congested, or if the road is too narrow and should be increased with lanes.
  • Si represents the lane change failure rate of the target vehicle
  • S f represents the failure of the target vehicle to change lanes
  • S s represents the successful lane change of the target vehicle.
  • the lane change failure rate of all target vehicles in the target area within any time range can also be calculated.
  • the road transportation management department can adopt appropriate improvement measures to solve the traffic problem and further improve the target vehicle's distance from the target vehicle's location. The efficiency of changing lanes from the area to the target area.
  • the present disclosure also provides a specific implementation manner of the traffic area lock control device. See the examples below.
  • the traffic area lock control device provided by the embodiment of the present disclosure includes:
  • the acquisition module 901 is used to acquire the multi-dimensional spatiotemporal position of the target vehicle
  • the determination module 902 is used to determine the area where the target vehicle is located based on the multi-dimensional spatio-temporal position
  • the determination module 902 is also used to determine the offset area based on the projection area and the preset offset amount of the area where the target vehicle is located;
  • the determination module 902 is also used to determine the target area according to the projection area and the offset area;
  • the control module 903 is used to control the area lock in the target area to be occupied when the area where the target vehicle is located is in a neighboring area of the target area and the area lock of the target area is in an idle state, and control the target vehicle to move from the area where the target vehicle is located.
  • the area changes lanes to the target area.
  • the target area of the target vehicle is determined, the area lock in the target area is controlled to change from the idle state to the occupied state, and the target vehicle is controlled to change lanes from the area where the target vehicle is located to the target area, so that the The target area is only for target vehicles to change lanes, which can avoid traffic accidents caused by multiple vehicles occupying the same position in the same lane from multiple different lanes.
  • the determination module 902 may include:
  • the acquisition unit is used to acquire the projection points of the target vehicle in multiple preset directions according to the multi-dimensional space-time position;
  • the determining unit is used to determine the area according to the projection point.
  • the projection area of the area can also be obtained through positioning methods such as grid division or GPS.
  • positioning methods such as grid division or GPS.
  • errors may occur due to inaccuracies due to too large grids, poor GPS signals, or long distances. Big difference problem.
  • the determination unit in the determination module 902 can also be used to:
  • the traffic area lock control device provided by the embodiment of the present disclosure may also include:
  • the receiving module 1001 is configured to receive an area lock request from the target vehicle.
  • the area lock request is an area lock request sent by the target vehicle based on the vehicle information of the target vehicle and the target area;
  • Adding module 1002 is used to add the region lock request to the region lock request list of the target region.
  • the control module 903 is also configured to control the target vehicle to change lanes from the area where the target vehicle is located to the target when it is detected that the area lock of the target area is in an idle state and the area lock request of the target vehicle is reached according to the order in the area lock request list. area.
  • the area lock request also includes the vehicle type of the target vehicle, as shown in Figure 10, the determination module 902 is also used to determine the priority of the target vehicle based on the preset correspondence between the vehicle type and the priority when receiving the area lock request of the target vehicle.
  • the adding module 1002 is also used to add the region lock request to the region lock request list based on the priority.
  • the acquisition module is also used to obtain the first request time, the first The request time is the request time when the region lock request is added to the region lock request list.
  • the acquisition module is also used to obtain the second request time, which is the request time when the area lock request of the vehicle currently occupying the target area is added to the area lock request list.
  • the traffic area lock control device may also include: an obtaining module 1003, used to calculate the difference between the first request time and the second request time, and obtain the waiting time of the target vehicle.
  • the control module 903 is also used to control the target vehicle to change lanes from the area where the target vehicle is located to the target area after waiting for a period of time.
  • the acquisition module 1003 is also used to:
  • the target area is adjusted to obtain the adjusted target area.
  • control module 903 is also used to:
  • the area lock in the adjusted target area is controlled to be in the occupied state, and the target vehicle is controlled to move from the target area.
  • the area where the vehicle is located changes lanes to the adjusted target area.
  • the obtaining unit 1003 includes:
  • the acquisition unit is used to obtain the traffic accident probability values, vehicle density and initialization parameters of the target area corresponding to the target vehicles in all lanes;
  • Obtain unit used to obtain the initialization parameter matrix based on the initialization parameters
  • An update unit used to update the initialization parameter matrix based on the second traffic accident probability threshold based on the Sparrow search algorithm
  • the output unit is used to substitute the updated initialization parameter matrix into the machine learning algorithm calculation.
  • the updated initialization parameter matrix satisfies the multi-objective function optimization model, output the adjusted weighted parameters corresponding to the target areas corresponding to the target vehicles in all lanes;
  • the training unit is used to train the optimal solution model based on the adjusted parameters corresponding to the target areas corresponding to the vehicles in all lanes.
  • Figure 11 shows a schematic diagram of the hardware structure of traffic area lock control provided by an embodiment of the present disclosure.
  • the traffic area lock control device may include a processor 1101 and a memory 1102 storing computer program instructions.
  • the above-mentioned processor 1101 may include a central processing unit (CPU), or an Application Specific Integrated Circuit (ASIC), or one or more integrated circuits that may be configured to implement embodiments of the present disclosure.
  • CPU central processing unit
  • ASIC Application Specific Integrated Circuit
  • Memory 1102 may include bulk storage for data or instructions.
  • the memory 1102 may include a hard disk drive (HDD), a floppy disk drive, Flash memory, optical disk, magneto-optical disk, tape or Universal Serial Bus (USB) drive or a combination of two or more of these.
  • Memory 1102 may include removable or non-removable (or fixed) media, where appropriate. Where appropriate, memory 1102 may be internal or external to the traffic area lock control device. In certain embodiments, memory 1102 is non-volatile solid-state memory.
  • memory 1102 may include read-only memory (ROM), random access memory (RAM), magnetic disk storage media devices, optical storage media devices, flash memory devices, electrical, optical, or other physical/tangible memory storage devices .
  • ROM read-only memory
  • RAM random access memory
  • magnetic disk storage media devices e.g., magnetic disks
  • optical storage media devices e.g., magnetic disks
  • flash memory devices e.g., electrical, optical, or other physical/tangible memory storage devices.
  • memory 1102 includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by a or a plurality of processors), which is operable to perform the operations described with reference to the method according to the first aspect of the present disclosure.
  • the processor 1101 reads and executes the computer program instructions stored in the memory 1102 to implement any of the traffic area lock control methods in the above embodiments.
  • the traffic zone lock control device may further include a communication interface 1103 and a bus 1110. Among them, as shown in Figure 11, the processor 1101, the memory 1102, and the communication interface 1103 are connected through the bus 1110 and complete communication with each other.
  • the communication interface 1103 is mainly used to implement communication between various modules, devices, units and/or equipment in the embodiments of the present disclosure.
  • Bus 1110 includes hardware, software, or both, coupling the components of the traffic zone lock control device to each other.
  • the bus may include Accelerated Graphics Port (AGP) or other graphics bus, Enhanced Industry Standard Architecture (EISA) bus, Front Side Bus (FSB), HyperTransport (HT) interconnect, Industry Standard Architecture (ISA) Bus, Infinite Bandwidth Interconnect, Low Pin Count (LPC) Bus, Memory Bus, Micro Channel Architecture (MCA) Bus, Peripheral Component Interconnect (PCI) Bus, PCI-Express (PCI-X) Bus, Serial Advanced Technology Attachment (SATA) bus, Video Electronics Standards Association Local (VLB) bus or other suitable bus or a combination of two or more of these.
  • bus 1110 may include one or more buses.
  • the embodiment of the present disclosure can provide a computer storage medium for implementation.
  • the computer storage medium stores computer program instructions; When the computer program instructions are executed by the processor, any one of the traffic area lock control methods in the above embodiments is implemented.
  • the functional blocks shown in the above structural block diagram can be implemented as hardware, software, firmware or a combination thereof.
  • it may be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, or the like.
  • ASIC application specific integrated circuit
  • elements of the disclosure are programs or code segments that are used to perform required tasks.
  • the program or code segments may be stored in a machine-readable medium or transmitted over a transmission medium or communications link via a data signal carried in a carrier wave.
  • "Machine-readable medium" may include any medium capable of storing or transmitting information.
  • machine-readable media examples include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, and the like.
  • Code segments may be downloaded via computer networks such as the Internet, intranets, and the like.
  • Such a processor may be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It will also be understood that each block in the block diagrams and/or flowchart illustrations, and combinations of blocks in the block diagrams and/or flowchart illustrations, can also be implemented by special purpose hardware that performs the specified functions or actions, or can be implemented by special purpose hardware and computer A combination of instructions is implemented.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Control Of Driving Devices And Active Controlling Of Vehicle (AREA)
  • Traffic Control Systems (AREA)

Abstract

本公开公开了一种交通区域锁控制方法、装置、设备和存储介质，该方法包括：获取目标车辆的多维时空位置；根据多维时空位置，确定目标车辆所在的区域；根据目标车辆所在的区域的投影区域和预设偏移量，确定偏移区域；根据投影区域和偏移区域，确定目标区域；当目标车辆所在的区域处于目标区域的邻区，且目标区域的区域锁为空闲状态时，控制目标区域内的区域锁为占用状态，并控制目标车辆从目标车辆所在的区域变道至目标区域。

Description

交通区域锁控制方法、装置、设备和存储介质
相关申请的交叉引用
本公开主张在2022年08月17日在中国提交的中国专利申请号No.202210987081.3的优先权，其全部内容通过引用包含于此。
技术领域
本公开属于自动驾驶技术领域，尤其涉及一种交通区域锁控制方法、装置、设备和存储介质。
背景技术
基于5G边缘计算的自动驾驶系统，通过车辆配备的传感器测量周围车辆的距离，直接进行判断是否可以变道或者超车，经常出现多个车辆从多个不同的车道抢占同一车道相同位置导致的交通事故的问题。
发明内容
本公开实施例提供一种交通区域锁控制方法、装置、设备和存储介质，能够避免经常出现多个车辆从多个不同的车道抢占同一车道相同位置导致的交通事故。
第一方面，本公开实施例提供一种交通区域锁控制方法，方法包括：
获取目标车辆的多维时空位置；
根据多维时空位置，确定目标车辆所在的区域；
根据目标车辆所在的区域的投影区域和预设偏移量，确定偏移区域；
根据投影区域和偏移区域，确定目标区域；
当目标车辆所在的区域处于目标区域的邻区，且目标区域的区域锁为空闲状态时，控制目标区域内的区域锁为占用状态，并控制目标车辆从目标车辆所在的区域变道至目标区域。
第二方面，本公开实施例提供了一种交通区域锁控制装置，装置包括：
获取模块，用于获取目标车辆的多维时空位置；
确定模块，用于根据多维时空位置，确定目标车辆所在的区域；
确定模块，还用于根据目标车辆所在的区域的投影区域和预设偏移量，确定偏移区域；
确定模块，还用于根据投影区域和偏移区域，确定目标区域；
控制模块，用于当目标车辆所在的区域处于目标区域的邻区，且目标区域的区域锁为空闲状态时，控制目标区域内的区域锁为占用状态，并控制目标车辆从目标车辆所在的区域变道至目标区域。
第三方面，本公开实施例提供了一种交通区域锁控制设备，设备包括：处理器以及存储有计算机程序指令的存储器；
处理器执行计算机程序指令时实现如第一方面所述的方法。
第四方面，本公开实施例提供了一种计算机存储介质，计算机可读存储介质上存储有计算机程序指令，计算机程序指令被处理器执行时实现如第一方面所述的方法。
第五方面，本公开实施例提供了一种计算机程序产品，其特征在于，计算机程序产品中的指令由电子设备的处理器执行时，使得电子设备执行如第一方面所述的方法。
本公开实施例提供的交通区域锁控制方法、装置、设备和存储介质，能够通过获取目标车辆的多维时空位置；再根据多维时空位置，确定目标车辆所在的区域；再根据目标车辆所在的区域的投影区域和预设偏移量，确定偏移区域；再根据投影区域和偏移区域，确定目标区域；当目标车辆所在的区域处于目标区域的邻区，且目标区域的区域锁为空闲状态时，控制目标区域内的区域锁为占用状态，并控制目标车辆从目标车辆所在的区域变道至目标区域。如此，根据目标车辆的多维时空位置，确定目标车辆的目标区域，控制目标区域内的区域锁从空闲状态变更为占用状态，并控制目标车辆从目标车辆所在的区域变道至目标区域，使得该目标区域仅供目标车辆变道，能够避免多个车辆从多个不同的车道抢占同一车道相同位置导致的交通事故。
附图说明
为了更清楚地说明本公开实施例的技术方案，下面将对本公开实施例中 所需要使用的附图作简单的介绍，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图：
图1示出了本公开实施例提供的基于移动边缘计算节点控制自动驾驶车辆进行变道的示意图；
图2示出了本公开实施例提供的将自动驾驶车辆作为用户设备注册接入移动边缘计算节点的流程示意图；
图3示出了本公开实施例提供的自动驾驶车辆中配备的传感器传送自动驾驶车辆驾驶数据至移动边缘计算节点用户面功能模块的示意图；
图4是本公开实施例提供的交通区域锁控制方法的流程示意图；
图5示出了本公开实施例提供的基于目标车辆上任意一个外部传感器获取部分投影区域的示意图；
图6示出了本公开实施例提供的基于多维时空位置进行平面投影的示意图；
图7示出了本公开实施例提供的区域锁请求列表的示意图；
图8示出了本公开实施例提供的最优解求解模型的训练方法的流程示意图；
图9是本公开实施例提供的一种交通区域锁控制装置的结构示意图；
图10是本公开实施例提供的另一种交通区域锁控制装置的结构示意图；
图11是本公开实施例提供的交通区域锁控制设备的结构示意图。
具体实施方式
下面将详细描述本公开的各个方面的特征和示例性实施例，为了使本公开的目的、技术方案及优点更加清楚明白，以下结合附图及具体实施例，对本公开进行进一步详细描述。应理解，此处所描述的具体实施例仅意在解释本公开，而不是限定本公开。对于本领域技术人员来说，本公开可以在不需要这些具体细节中的一些细节的情况下实施。下面对实施例的描述仅仅是为了通过示出本公开的示例来提供对本公开更好的理解。
需要说明的是，在本文中，诸如第一和第二等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来，而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。而且，术语“包括”、 “包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者设备不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者设备所固有的要素。在没有更多限制的情况下，由语句“包括……”限定的要素，并不排除在包括所述要素的过程、方法、物品或者设备中还存在另外的相同要素。
第五代移动通信技术(5th Generation Mobile Communication Technology，5G)是对网络技术的创新和变革。5G通信网络更加去中心化，在网络边缘部署小规模或者便携式数据中心，将终端请求的一部分业务交由本地化计算处理，以满足超可靠低时延通信和海量大规模连接物联网的超低时延需求。
移动边缘计算(Mobile Edge Computing，MEC)是5G的核心技术之一。移动边缘计算运行于网络边缘，通常具有较高的计算能力，尤其适合分析处理大量数据的计算任务。由于在地理位置上移动边缘计算节点距离用户及信息源非常邻近，使得MEC节点的网络响应用户请求的时延大大减小，降低了传输网和核心网发生网络拥塞的可能性，实现高可靠低时延的传输要求。
国际电信联盟无线通信局(International Tecommunication Union-Radiocommunica，ITU-R)定义了5G的三种典型应用场景，分别是增强型移动宽带(Enhanced Mobile Broadband，eMBB)、超可靠低时延通信(Ultra Reliable Low Latency Communication，uRLLC)和海量大规模连接物联网(massive Machine Type of Communication，mMTC)。
其中，eMBB主要面向如虚拟现实(Virtual Reality，VR)、增强现实(Augmented Reality，AR)和在线高清视频(4K/8K)等高带宽需求业务；uRLLC主要面向如智慧交通和智慧医疗等垂直行业的时延敏感的极端性能需求业务；mMTC主要面向智慧城市、智慧工业和智能水表等超千亿设备的高连接密度需求的物联网业务。
就智慧交通而言，智慧交通中的自动驾驶领域，主要是自动驾驶车辆与移动边缘计算相互协同工作，在各种复杂的交通环境中，经过智能综合地感知、判断、推理和决策后，实现对自动驾驶车辆进行控制，像人控制车辆一样。
具体地，自动驾驶车辆上搭载的自动驾驶系统通过车辆配备中内部和外部的传感器，获取自身状态及周边交通环境信息。其中，内部传感器主要包 括车速传感器、加速传感器、轮速传感器以及横摆角速度传感器等，外部传感器包括摄像头、激光雷达、毫米波雷达以及定位系统等。通过这些传感器收集自动驾驶车辆全方位的交通环境信息，并将自动驾驶车辆全方位的交通环境信息发送到MEC节点，由MEC节点进行相关的交通调度计算，随后，MEC节点将计算结果返回给自动驾驶系统，自动驾驶系统执行决策命令，控制自动驾驶车辆正常行驶，实现既保障交通顺畅有序，又避免发生交通堵塞或事故的效果。类似于人类驾驶员在驾驶过程中通过视觉、听觉和触觉等感官系统感知行驶环境和车辆状态。
现有的基于移动边缘计算，面向智慧交通方面的研究，大多数侧重于自动驾驶方面超低时延的任务分配、调度、迁移以及卸载等方向的研究。然而，当前基于移动边缘计算，对于避免交通事故的研究较少。
目前，MEC节点主要用于收集自动驾驶车辆周边的交通环境信息，并将这些交通环境信息反馈给自动驾驶车辆的自动驾驶系统，供辅助决策使用，并没有合理的综合利用这些交通环境信息，进行深度数据挖掘分析，真正起到智能计算调度中心的作用。基于此，在智慧交通中，出现每辆车的自动驾驶系统局部独立的判断而不考虑整体全局情况，容易造成多个车辆争抢同一车道相同位置而引发交通事故的问题。
当前，避免交通事故的主要技术手段是通过传感器测量周围车辆的距离，直接进行判断是否可以变道或者超车，主要适用于处理单双车道、比较空闲和车速较慢的简单低密度的交通场景，但无法胜任处理较多车道、比较拥堵和车速较快的复杂高密度的交通场景，并不能避免多个车辆从多个不同的车道抢占同一车道相同位置导致的交通事故的情况，不能满足智慧交通的实际需求。
为了解决相关技术问题，本公开实施例提供了一种交通区域锁控制方法、装置、设备和存储介质。在对本公开实施例提供的交通区域锁控制方法进行说明之前，先对本公开实施例所提供的基于MEC节点控制车辆进行变道的过程进行概括介绍。
如图1所示，云计算中心节点和多个起到交通区域锁控制调度中心作用的MEC节点建立通信，MEC节点与多个车道中的每一台自动驾驶车辆的自动驾驶系统建立通信。当自动驾驶车辆A和自动驾驶车辆B从车辆所在车道 变道至目标车道的目标区域R1，和/或自动驾驶车辆C和自动驾驶车辆D从车辆所在车道变道至目标车道的目标区域R2时，MEC节点控制自动驾驶车辆A和自动驾驶车辆B中任意一辆车辆从车辆所在车道变道至目标车道的目标区域R1，和/或自动驾驶车辆C和自动驾驶车辆D中任意一辆自动驾驶车辆从车辆所在车道变道至目标车道的目标区域R2。
为实现基于MEC节点控制自动驾驶车辆进行变道，需将自动驾驶车辆作为用户设备注册接入MEC节点。如图2所示，在基于MEC节点控制自动驾驶车辆进行变道之前，在5G网络信号覆盖范围内，需将自动驾驶车辆作为用户设备(User Equipment，UE)注册接入MEC节点。
将自动驾驶车辆作为UE注册接入MEC节点的具体过程如下：
1、UE发起注册请求，MEC节点所在的无线接入网(Radio Access Network，RAN)接收注册请求，并将注册请求转发至接入与移动性管理功能模块(Access and Mobility Management Function，AMF)。
2、AMF基于本地配置或者基于网络切片进行选择转发。
3、当初始的AMF判断不能为UE提供服务时，将发起AMF重定位；AMF收到注册请求，根据用户隐藏标识符(Subscription Concealed Identifier，SUCI)的路由选择码找到对应的认证服务器功能模块(Authentication Server Function，AUSF)提出鉴权请求。
4、AUSF返回鉴权参数，鉴权参数包括随机数据Random Number、认证数据Authentication Token以及期望的响应数据。
5、AMF将鉴权参数中的随机数据和认证数据发送给UE；UE使用认证数据对网络进行鉴权，然后返回给AMF响应数据；AMF对比响应数据，如果符合，则鉴权成功，再将响应数据发送给AUSF；AUSF对比响应数据，如果符合，则返回给AMF对应SUCI的用户永久标识符(Subscription Perpetual Identifier，SUPI)。
6、AMF发起至统一数据管理模块(Unified Data Management，UDM)进行登记，并在UDM进行签约数据管理的订阅。
7a、如果UDM中没有找到对应UE的初始注册信息，或最近一次注册已经更换了AMF，则AMF需要再向UDM进行终端连接管理的注册。
7b、AMF以国际移动用户识别码(International Mobile Subscriber  Identification Number，IMSI)参数的方式向UDM发起签约数据的请求，UDM返回IMSI的签约数据。
7c、AMF根据IMSI向UDM发起终端签约数据变化的订阅，UDM返回订阅成功的通知。
8、AMF将依据SUPI选择一个策略控制功能(Policy Control Function，PCF)。
9、在注册期间，AMF也可以根据移动性限制的情况向PCF进行关联策略的调整。
10、AMF向UE返回接受注册的消息。
11、UE回应完成注册接入5G网络MEC。
为实现基于MEC节点控制自动驾驶车辆进行变道，还需自动驾驶车辆中配备的传感器传送自动驾驶车辆驾驶的数据到MEC节点中的用户面功能模块(User Plane Function，UPF)。
如图3所示，自动驾驶车辆中配备的内部和外部传感器，获取自身状态及周边环境信息，传感器再将收集到的车辆全方位的交通环境信息，通过自动驾驶系统传送到MEC节点中的UPF，UPF负责数据传输和数据处理。
接下来，对本公开实施例所提供的交通区域锁控制方法进行介绍。
图4是本公开实施例提供的交通区域锁控制方法的流程示意图。如图4所示，本公开实施例提供的交通区域锁控制方法包括以下步骤：
S401、获取目标车辆的多维时空位置。
目标车辆为受MEC节点控制，将根据区域锁的状态，从目标车辆所在的车道变道至区域锁所在的车道的车辆。区域锁，即锁定目标区域的交通区域锁。目标车辆的多维时空位置，根据目标车辆的多维时空数据确定。
多维时空数据包括但不限于目标车辆行驶的当前时刻和在该时刻下目标车辆的纬度、经度、高度、行驶方向以及所在单位面积内的地理区域的车辆数量。
需说明，为便于对交通区域锁控制方法进行说明，将自动驾驶车辆A作为目标车辆进行说明。
MEC节点，通过自动驾驶系统，获取目标车辆中配备的内部和外部传感器收集的目标车辆的多维时空数据，以便根据目标车辆的多维时空数据，确 定目标车辆的多维时空位置，再根据目标车辆的多维时空位置，确定目标车辆所在的区域。
S402、根据多维时空位置，确定目标车辆所在的区域。
目标车辆所在的区域包括目标车辆占据目标车辆所在车道的区域。目标车辆所在的区域包括多维时空位置。确定目标车辆所在的区域，以便于在MEC节点根据目标车辆所在的区域，确定目标车辆的目标区域。
在一个实施例中，可以根据多维时空位置，任意圈定目标车辆占据目标车辆所在车道的区域，作为目标车辆所在的区域。
在一个实施例中，还可以基于目标车辆的车型尺寸，圈定目标车辆占据目标车辆所在车道的区域，作为目标车辆所在的区域。
在本公开的一些实施例中，为了进一步精确目标车辆所在的区域，以便提高控制目标车辆从目标车辆所在的区域变道至目标区域的效率，步骤S402还可以包括步骤S4021至步骤S4022。
S4021：根据多维时空位置，获取目标车辆多个预设方向的投影点位。
S4022：根据投影点位，确定所述区域。
预设方向包括目标车辆左前方、左后方、右前方和右后方。在目标车辆左前方、左后方、右前方和右后方中任意一个方向，至少配备1个以上外部传感器。通过目标车辆的外部传感器，获取目标车辆外围左前方、左后方、右前方和右后方至少四个投影点位的数据，再根据投影点位的数据，计算出目标车辆占据目标车辆所在车道的平面投影区域，将该平面投影区域作为目标车辆所在的区域。
可选的，在一些实施例中，外部传感器可以是激光雷达或毫米波雷达等类型的传感器，外部传感器能够全角度发射激光、毫米波等电磁波束。每个车道两边均安装感应带，感应带可以准确感应并接收目标车辆发射的电磁波束的每个投影点位。
基于目标车辆多个预设方向的投影点位，计算出目标车辆占据目标车辆所在车道的平面投影区域，并将该平面投影区域作为目标车辆所在的区域，实现了进一步精确目标车辆所在的区域，以便提高控制目标车辆从目标车辆所在的区域变道至目标区域的效率的效果。
S403、根据目标车辆所在的区域的投影区域和预设偏移量，确定偏移区 域。
目标车辆所在的区域的投影区域，为基于投影的位置关系，目标车辆占据目标车辆所在车道的平面投影区域，投影到目标车辆将要变道的车道的投影区域。
预设偏移量为目标车辆向计划变道区域变道时，预先设定的目标车辆所在的区域的投影区域的偏移量。其中，目标车辆所在的区域的投影区域，即目标车辆的多维时空位置的平面投影。根据投影区域和预设偏移量，计算出偏移区域。确定偏移区域，以便目标车辆在变道时，伴随目标车辆的行驶过程，目标车辆的目标区域中预留出目标车辆与目标车辆的前后车辆之间的安全距离，以免目标车辆与目标车辆的前后车辆之间发生追尾事件。
在一个实施例中，预设偏移量可以包括目标车辆的多维时空位置的平面投影的前方偏移量和后方偏移量。对应的，偏移区域包括前方偏移区域和后方偏移区域。根据区域的投影区域和前方偏移量，确定前方偏移区域，根据区域的投影区域和后方偏移量，确定后方偏移区域。
在本公开的一些实施例中，通过划分网格或GPS等定位方法，也可以获得区域的投影区域，但会出现因网格过大不准确、GPS信号差或者距离远误差大的问题。为了进一步精确偏移区域，以便提高控制目标车辆从目标车辆所在的区域变道至目标区域的效率，步骤S403还可以包括步骤S4031至步骤S4032。
S4031：根据目标车辆所在的区域的投影的位置关系，确定投影区域。
S4032：根据预设偏移量和投影区域，确定偏移区域。
如图5所示，目标车辆所在的区域的投影区域包括将目标车辆的一个外部传感器作为投影点位，横扫至少90度角度，发射激光或毫米波等电磁波束，向计划变道区域两边的感应带进行平面投影，形成的投影区域。
基于此，根据目标车辆外围左前方、左后方、右前方和右后方至少四个投影点位的坐标，向计划变道区域两边的感应带进行平面投影，形成目标车辆所在的区域的投影区域。
上述通过目标车辆所在的区域的投影的位置关系，获取目标车辆的投影区域的方法，依据激光或毫米波等波长较短的电磁波传播接近直线传播的原理，相比划分网格或GPS等定位方法，能够更加迅速、直观且准确的获取目 标车辆的投影区域，避免了因网格过大不准确、GPS信号差或者距离远误差大的问题。
根据预设偏移量中的前方偏移量，将投影区域向前方偏移扩大，偏移扩大的区域为前方偏移区域。或者，根据预设偏移量中的后方偏移量扩大，将投影区域向后方偏移，偏移扩大的区域为后方偏移区域。
通过区域的投影的位置关系，获取目标车辆的投影区域，并对投影区域进行偏移扩大，确定偏移区域，能够进一步精确偏移区域，以便提高控制目标车辆从目标车辆所在的区域变道至目标区域的效率。
S404、根据投影区域和偏移区域，确定目标区域。
图6示出了本公开实施例提供的基于多维时空位置进行平面投影的示意图。如图6所示，目标区域包括投影区域和偏移区域，基于多维时空位置进行平面投影，根据投影区域和偏移区域，可以确定目标区域。
具体地，多维时空位置、多维时空位置的平面投影、目标车辆所在的区域以及目标区域的关系表示如下：
其中，R′表示目标车辆所在的区域；R表示目标车辆的目标区域；ti表示时间；Gi表示在ti时刻，目标车辆的多维时空位置；表示在ti时刻，目标车辆的多维时空位置的平面投影；wi表示目标车辆的纬度；li表示目标车辆的经度；hi表示目标车辆的高度；di表示目标车辆的行驶方向；ci表示目标车辆的所在单位面积内的地理区域的车辆数量。
目标区域表示如下：
其中，R表示目标车辆的目标区域；γ表示加权参数，可以成倍数调整目标区域R扩大；ti表示时间；表示在时间i范围内，目标车辆的多维时空位置的平面投影；wi表示在时间i范围内，目标车辆的纬度；li表示在时间i范围内，目标车辆的经度；hi表示在时间i范围内，目标车辆的高度；di表示在时间i范围内，目标车辆的行驶方向；ci表示在时间i范围内，目标车辆的所在单位面积内的地理区域的车辆数量；表示在时间i范围内，目标车辆的多维时空位置的平面投影的前方偏移区域；表示在时间i范围内，目标车辆的多维时空位置的平面投影的后方偏移区域；α表示加权参数，可以调整前方偏移区 域β表示加权参数，可以调整后方偏移区域初始化时，即在ti＝0时刻，γ、α和β可以为1。
前方偏移量和后方偏移量随着目标车辆的车速的提高而增加。例如，车速低于20km/h，偏移量保持10米；车速低于40km/h，偏移量保持40米；车速低于60km/h，偏移量保持60米；车速低于80km/h，偏移量保持80米；车速低于100km/h，偏移量保持150米到200米。
可以预先设定，前、后方偏移区域为目标车辆的前、后方偏移量(即距离)与道路宽度的乘积，目标车辆的前、后方偏移量至少为目标车辆所在车道最大允许车速与时间的乘积。
目标车辆的前、后方偏移量分别与目标车辆所在车道最大允许车速与时间的乘积的关系表示如下：
其中，ti表示时间；GF表示在时间i范围内，目标车辆的前方偏移量；表示在时间i范围内，目标车辆的后方偏移量；Vmax表示目标车辆所在车道最大允许车速。
目标车辆所在的区域和目标区域，是根据目标车辆的多维时空位置确定的在时间范围内动态对应的区域。
S405、当目标车辆所在的区域处于目标区域的邻区，且目标区域的区域锁为空闲状态时，控制目标区域内的区域锁为占用状态，并控制目标车辆从目标车辆所在的区域变道至目标区域。
目标区域的区域锁为空闲状态，表示目标区域内不存在区域锁；目标区域内的区域锁为占用状态，表示目标区域内存在区域锁。
控制目标区域内的区域锁为占用状态，表示MEC节点将目标车辆的区域锁添加到目标区域内，使得目标区域内存在目标车辆的区域锁。
当目标车辆所在的区域处于目标区域的邻区时，目标车辆所在的区域的投影区域在目标区域的范围内。基于MEC节点监测到目标车辆所在的区域处于目标区域的邻区，且目标区域的区域锁为空闲状态，MEC节点将目标车辆的区域锁添加到目标区域内，使得目标区域内存在目标车辆的区域锁，并控制目标车辆从目标车辆所在的区域变道至目标区域。
通过获取目标车辆的多维时空位置；再根据多维时空位置，确定目标车 辆所在的区域；再根据目标车辆所在的区域的投影区域和预设偏移量，确定偏移区域；再根据投影区域和偏移区域，确定目标区域；当目标车辆所在的区域处于目标区域的邻区，且目标区域的区域锁为空闲状态时，控制目标区域内的区域锁为占用状态，并控制目标车辆从目标车辆所在的区域变道至目标区域，如此，根据目标车辆的多维时空位置，确定目标车辆的目标区域，控制目标区域内的区域锁从空闲状态变更为占用状态，并控制目标车辆从目标车辆所在的区域变道至目标区域，使得该目标区域仅供目标车辆变道，能够避免多个车辆从多个不同的车道抢占同一车道相同位置导致的交通事故。
在本公开的一些实施例中，为了精准处理在较多车道、比较拥堵和车速较快的复杂高密度的交通场景之下，出现多个车辆从多个不同的车道抢占同一车道相同位置导致的交通事故的问题，在步骤S401之前，还可以包括步骤S406至步骤S407。
S406：获取目标车辆所在单位面积内的地理区域的车辆密集程度和交通事故发生概率。
S407：当车辆密集程度超过车辆密集程度阈值，和/或交通事故发生概率超过第一交通事故发生概率阈值时，获取目标车辆的多维时空位置。
根据目标车辆所在单位面积内的地理区域和在经过该地理区域的所有车辆数量，建立多维时空车辆密集程度模型。
多维时空车辆密集程度模型表示如下：
其中，M(ci)表示目标车辆所在单位面积内的地理区域的车辆密集程度；ci表示在时间i范围内，目标车辆所在单位面积内的地理区域的车辆数量；X表示目标车辆的所在单位面积内的地理区域面积的大小。
根据目标车辆所在单位面积内的地理区域发生交通事故的车辆数量和经过该地理区域的所有车辆数量，即在该交通区域发生事故的历史统计数据，建立交通事故预警模型。
交通事故预警模型表示如下：
其中，P(hi)表示在时间i范围内，目标车辆所在单位面积内的地理区域的交通事故发生概率；hi表示在时间i范围内，目标车辆所在单位面积内的地理 区域发生交通事故的车辆数量；ci表示在时间i范围内，目标车辆所在单位面积内的地理区域的车辆数量。
通过多维时空车辆密集程度模型，获取目标车辆所在单位面积内的地理区域的车辆密集程度。通过交通事故预警模型，获取目标车辆所在单位面积内的地理区域的交通事故发生概率。其中，单位面积可以预先设定，并根据实际需求进行调整。
当车辆密集程度超过车辆密集程度阈值，或交通事故发生概率超过第一交通事故发生概率阈值，或车辆密集程度超过车辆密集程度阈值且交通事故发生概率超过第一交通事故发生概率阈值时，表示目标车辆所在单位面积内的地理区域处于较多车道、比较拥堵和车速较快的复杂高密度的交通场景。
车辆密集程度与车辆密集程度阈值的关系和交通事故发生概率与第一交通事故发生概率阈值的关系表示如下：
其中，M(ci)表示目标车辆所在单位面积内的地理区域的车辆密集程度；Mθ表示车辆密集程度初始阈值；P(hi)表示在时间i范围内，目标车辆所在单位面积内的地理区域的交通事故发生概率；Pθ表示交通事故发生概率初始阈值。
在一个实施例中，当车辆密集程度超过车辆密集程度阈值，或交通事故发生概率超过第一交通事故发生概率阈值，或车辆密集程度超过车辆密集程度阈值且交通事故发生概率超过第一交通事故发生概率阈值时，可以自动触发MEC节点启动，以便MEC节点监测目标车辆所在的区域是否处于目标区域的邻区，且目标区域的区域锁为空闲状态，控制目标区域内的区域锁为被目标车辆所占有的占用状态，并控制目标车辆从目标车辆所在的区域变道至目标区域。
在一个实施例中，当车辆密集程度未超过车辆密集程度阈值，或交通事故发生概率未超过第一交通事故发生概率阈值，或车辆密集程度未超过车辆密集程度阈值且交通事故发生概率未超过第一交通事故发生概率阈值时，可以不触发MEC节点启动，方便用户灵活控制目标车辆行驶。
在一个实施例中，可以结合公式(4)和公式(5)计算出的车辆密集程度和交通事故概率，对需要锁定的目标区域范围进行优化，如果达到公式(6) 的条件，则通过调整公式(2)中的加权参数，适当扩大目标区域范围，即区域锁锁定的范围，从而降低车辆密集程度和交通事故概率。
当车辆密集程度超过车辆密集程度阈值，或交通事故发生概率超过第一交通事故发生概率阈值，或车辆密集程度超过车辆密集程度阈值且交通事故发生概率超过第一交通事故发生概率阈值时，自动触发MEC节点启动，可以精准处理在较多车道、比较拥堵和车速较快的复杂高密度的交通场景之下，出现多个车辆从多个不同的车道抢占同一车道相同位置导致的交通事故的问题。
在本公开的一些实施例中，为了进一步避免多个车辆从多个不同的车道抢占同一车道相同位置导致的交通事故的情况，当目标区域内的区域锁为占用状态时，本公开实施例提供的交通区域锁控制方法还可以包括步骤S501至步骤503。
S501：接收目标车辆的区域锁请求，区域锁请求为目标车辆基于目标车辆的车辆信息和目标区域发送的区域锁请求。
S502：将区域锁请求添加至目标区域的区域锁请求列表。
S503：当检测到目标区域的区域锁为空闲状态，且按照区域锁请求列表中的顺序达到目标车辆的区域锁请求时，控制目标车辆从目标车辆所在的区域变道至目标区域。
当目标区域内的区域锁为占用状态时，在目标车辆进行变道之前，目标车辆需先向MEC节点发送目标车辆基于目标车辆的车辆信息和目标区域发送的区域锁请求。例如，区域锁请求包括车牌号为XXX，请在所需占用的目标车道的目标区域R，增加区域锁K。
基于目标区域内存在其他的车辆的区域锁，MEC节点将通过队列模式，管理针对目标区域进行请求的区域锁，并将区域锁请求添加至目标区域的区域锁请求列表(即区域锁请求队列)中。
在一个实施例中，当目标区域内的区域锁为空闲状态时，即没有其他车辆对目标车道的目标区域拥有区域锁，或者其他车辆所请求的区域锁已经释放或失效，在目标车辆进行变道之前，MEC节点直接允许目标车辆对目标区域增加区域锁。
例如，在当前时刻ti，目标区域R为空闲状态，则MEC节点直接允许目 标车辆对目标区域增加区域锁Ki。
在一个实施例中，如图7所示，根据区域锁请求，MEC节点按照接收区域锁请求的时间先后顺序，将区域锁请求添加至目标区域的区域锁请求列表中。其中，区域锁请求列表包括正处于排队状态的区域锁K1、K2、K3、……、Ki,i＝1,2,3…。
当检测到目标区域的区域锁为空闲状态，且按照区域锁请求列表中的顺序达到目标车辆的区域锁请求，即MEC节点检测到目标区域不存在其他车辆的区域锁，且在目标车辆的区域锁之前的区域锁失效或释放后，MEC节点控制目标车辆从目标车辆所在的区域变道至目标区域。
通过将区域锁请求添加至目标区域的区域锁请求列表，并基于区域锁请求列表对目标区域内的区域锁进行管控，能进一步避免多个车辆从多个不同的车道抢占同一车道相同位置导致的交通事故的情况。
在一个实施例中，MEC节点控制目标车辆的区域锁的加锁过程。目标车辆计划占用指定的交通区域(即目标区域)时，基于目标车辆的多维时空位置，向MEC节点发送区域锁请求，并申请对目标区域进行加锁。MEC节点监测目标区域的区域锁是否为空闲状态，如果目标区域的区域锁是空闲状态，则将目标区域的区域锁更新为被目标车辆所占用的占用状态。随后，MEC节点返回响应成功的结果至目标车辆，并控制目标车辆驶入目标区域。
需注意，如果目标区域的区域锁是占用状态，则MEC节点将区域锁请求添加至目标区域的区域锁请求列表。其中，区域锁请求列表存在队列最大承受长度。MEC节点将区域锁请求添加至目标区域的区域锁请求列表之后，若当前的区域锁请求列表未超过队列最大承受长度，则MEC节点返回排队等待状态至目标车辆；若当前的区域锁请求列表超过队列最大承受长度，则MEC节点返回失败结果及原因至目标车辆的自动驾驶系统。
在一个实施例中，MEC节点控制目标车辆的区域锁的解锁过程。对区域锁解锁分为主动解锁模式和被动解锁模式。
主动解锁模式是由目标车辆主动向区域锁控制系统申请区域锁解锁请求，区域锁控制系统收到区域锁解锁请求后，对目标区域释放区域锁，并将目标区域的区域锁的标志位更新为空闲状态。主动解锁模式主要依赖目标车辆及时准确的根据自己当前的位置向区域锁控制系统申请解锁，需要目标车辆的 内外部传感器、自动驾驶系统以及所处的5G网络具备较高的灵敏度和性能。
被动解锁模式是在目标车辆驶离目标区域后，由区域锁控制系统通过5G NR厘米波或毫米波定位技术，自动对目标车辆的位置进行判断，当判断目标车辆已经驶离目标区域后，释放指定的区域锁，并将目标区域的区域锁的标志位更新为空闲状态。目标车辆被动接受区域锁控制系统处理结果。被动解锁模式可以有效弥补目标车辆的主动解锁模式能力不足的情况，即当目标车辆的内外部传感器、自动驾驶系统以及所处的5G网络并不具备较高的灵敏度和性能，区域锁控制系统将计算压力主要集中于移动边缘计算侧。
在实际应用中，可以采用主动解锁模式或被动解锁模式的任意一个或者相互结合方式实现区域快速解锁。
在一个实施例中，一般情况下，默认设置一辆目标车辆最多持有一个区域锁，锁粒度的大小，即区域锁的锁定的目标区域的面积，最小可以为目标车辆平面面积。
在一个实施例中，一辆目标车辆也可以申请多个区域锁，在设置目标车辆的区域锁时，可以控制目标车辆的区域锁的数量的设置。
在本公开的一些实施例中，为了优先处理110警车、120医疗救护车、119消防救火车等救援车辆进行变道，区域锁请求还包括目标车辆的车辆类型，本公开实施例提供的交通区域锁控制方法还可以包括步骤S601至步骤S602。
S601：当接收到目标车辆的区域锁请求时，基于预设的车辆类型与优先级的对应关系，确定目标车辆的优先级。
S602：基于优先级将区域锁请求添加至区域锁请求列表。
通过对区域锁加权，赋予紧急情况的目标车辆的区域锁不同的权值，在根据区域锁不同的权值分配给目标车辆不同的优先级，即预设的车辆类型不同，赋予不同预设的车辆类型的目标车辆不同的优先级。目标车辆的区域锁权值越大，目标车辆对应的交通情况越紧急，则目标车辆的优先级越高。
预设的车辆类型包括优先级车辆和常规车辆。优先级车辆包括但不限于110警车、120医疗救护车、119消防救火车等救援车辆。根据车辆的救援属性不同，还可以进一步对优先级车辆进行细分。例如，优先级从高到低，依次设定120医疗救护车、119消防救火车和110警车的优先级为一级、二级和三级。当120医疗救护车、119消防救火车和110警车同时发送区域锁请求时， 优先允许120医疗救护车进行变道，待120医疗救护车变道成功后，依次允许119消防救火车和110警车。
需要说明的是，在变道至目标区域时，任意一种优先级车辆的优先级高于常规车辆的优先级。
当接收到目标车辆的区域锁请求时，将优先级等级最高的车辆所对应的区域锁请求添加至区域所请求列表中最靠前的位置，待MEC节点检测到目标区域不存在其他车辆的区域锁，且在优先级等级最高的车辆的区域锁之前的区域锁失效或释放后，MEC节点控制优先级等级最高的车辆从车辆所在的区域变道至目标区域。
通过预设的车辆类型与优先级的对应关系，确定目标车辆的优先级，再基于优先级将区域锁请求添加至区域锁请求列表，可优先处理110警车、120医疗救护车、119消防救火车等救援车辆进行变道。
例如，如图1所示，自动驾驶车辆C为常规车辆，自动驾驶车辆D为优先级车辆。自动驾驶车辆C和自动驾驶车辆D从车辆所在车道变道至目标车道的目标区域R2时，由于自动驾驶车辆D的优先级高于自动驾驶车辆C，在自动驾驶车辆C和自动驾驶车辆D中，优先让自动驾驶车辆D从车辆所在车道变道至目标车道的目标区域R2。
再例如，在目标区域的区域锁为占用状态的情况下，当MEC节点接收到自动驾驶车辆C和自动驾驶车辆D发送的区域锁请求时，由于自动驾驶车辆D的优先级高于自动驾驶车辆C，在区域锁请求列表中，自动驾驶车辆D排列在自动驾驶车辆C前面。
对于多个车辆同时请求对同一车道相同位置的目标区域增加区域锁时，MEC节点根据预设的车辆类型，给不同的目标车辆分给不同的优先级区域锁，优先处理优先级车辆进行变道，可以保障救援车辆能够快速通过该路段，减少排队等待时间，避免因等待而导致错过最佳救援时间。
在一个实施例中，考虑到如果全部车辆的区域锁均分配不同的权值，可能面临区域锁权值较小的目标车辆一直在排队等待的问题。因此，针对除优先级车辆之外的其他常规车辆的区域锁，不区分优先级，而是按先请求先服务的原则，由MEC节点进行调度处理。
对于常规车辆发起的相同目标车道的相同目标区域的区域锁请求，采用 先请求先服务的队列处理逻辑。当目标车辆驶离目标车道的目标区域，则对应的队列头部的区域锁释放，顺移由下一个目标车辆的区域锁占用目标区域，那么下一个目标车辆可以选择驶入目标车道的目标区域。如果下一个目标车辆放弃驶入目标车道的目标区域，则该目标车辆的区域锁将失效，再依次由后面的目标车辆的区域锁占用目标区域，并且执行上述类似的过程。
基于此，也可以获悉区域锁具有独占排他性，即在该区域锁释放或失效之前，其他车辆不允许再对该目标车道的目标区域增加区域锁，需要排队等待。
在本公开的一些实施例中，为了便于用户了解变道所需要的等待的时长，丰富MEC节点所呈现的功能，在步骤S602之后，本公开实施例提供的交通区域锁控制方法还可以包括步骤S603至步骤S606。
步骤S603：获取第一请求时间，第一请求时间为区域锁请求添加至区域锁请求列表的请求时间。
步骤S604：获取第二请求时间，第二请求时间为当前占用目标区域的车辆的区域锁请求添加至区域锁请求列表的请求时间。
步骤S605：计算第一请求时间和第二请求时间的差值，获得目标车辆的等待时长。
步骤S606：控制目标车辆在等待时长后，从目标车辆所在的区域变道至目标区域。
等待时长为区域锁请求添加至区域锁请求列表的请求时间，距离当前占用目标区域的车辆的区域锁请求添加至区域锁请求列表的请求时间的时长。
基于第一请求时间和第二请求时间的差值，可计算出等待时长。但在实际行驶过程中，等待时长仍旧受目标区域的位置、目标区域的大小、当前车速、道路交通状况等多重因素影响。
等待时长表示如下：
Tw＝t1-t2   ……(7)
其中，Tw表示目标车辆从目标车辆所在的区域变道至目标区域的等待时长；t1表示区域锁请求添加至区域锁请求列表的请求时间；t2表示当前占用目标区域的车辆的区域锁请求添加至区域锁请求列表的请求时间。
计算等待时长，在便于用户了解变道所需要的等待的时长，丰富MEC 节点所呈现的功能的同时，也给用户提供了是否要变道的选择。例如，若用户认为等待时间过长，也可以人为选择主动解锁模式，放弃变道。
需要说明的是，上述交通区域锁控制方法涉及的加权参数γ、α和β、阈值Mθ和Pθ、车辆数量ci以及等待时长Tw，均可以通过机器学习算法不断的迭代计算进行调节，尽可能满足在目标车辆所在单位面积内的地理区域内，安全的容纳较多车辆、目标区域锁定范围小且准确以及等待时长缩短，避免产生交通拥堵现象。
在本公开的一些实施例中，为了提高目标车辆从目标车辆所在的区域变道至目标区域的效率，在步骤S404之后，本公开实施例提供的交通区域锁控制方法还可以包括步骤S701至步骤S702。
S701：将交通事故概率值、车辆密集程度和目标区域输入最优解求解模型，获得调整后的目标区域对应的加权参数；
S702：根据调整后的目标区域对应的加权参数，调整目标区域，获得调整后的目标区域。
现实场景中，一般期望能够满足在目标车辆所在单位面积内的地理区域内，安全的容纳较多车辆、目标区域锁定范围小且准确以及等待时长缩短。对应本公开实施例提供的交通区域锁控制方法中，则要求以自动驾驶发生交通事故的概率低、车辆密集程度高、锁定的目标区域小和等待时间短为目标。
基于此，将上述问题通过数据建模归结为多目标优化问题，建立多目标函数优化模型。将交通事故概率值、车辆密集程度和目标区域输入最优解求解模型，对多目标优化问题进行求解，获得满足多目标函数优化模型的最优解。
在一个实施例中，将交通事故概率值、车辆密集程度和目标区域输入最优解求解模型，输入最优解求解模型实时动态的调整目标区域对应的加权参数γ、α和β的数值，和/或阈值Mθ和Pθ以及车辆数量ci的数值，并迭代计算，获得满足多目标函数优化模型的调整后的目标区域对应的加权参数。
在一个实施例中，当MEC节点预测到较为拥堵的交通情况时，MEC节点通过最优解求解模型，将多目标函数优化模型作为迭代计算的停止条件，迭代计算获得最优加权参数γ、α和β、阈值Mθ和Pθ以及车辆数量ci的最优数值，实现调控交通，直至在未来的一段时间内交通恢复到正常。然后，结合计算 的最优数值计算得出交通区域锁定范围，即目标区域。
通过以自动驾驶发生交通事故的概率低、车辆密集程度高、锁定的目标区域小和等待时间短为目标，调整目标区域，可以提高目标车辆从目标车辆所在的区域变道至目标区域的效率。
在本公开的一些实施例中，为了进一步提高目标车辆从目标车辆所在的区域变道至目标区域的效率，在步骤S405中还可以包括步骤S4051。
S4051：当目标车辆所在的区域处于调整后的目标区域的邻区，且调整后的目标区域的区域锁为空闲状态时，控制调整后的目标区域内的区域锁为占用状态，并控制目标车辆从目标车辆所在的区域变道至调整后的目标区域。
基于调整后的目标区域进行变道，能够达到在目标车辆所在单位面积内的地理区域内，安全的容纳较多车辆、目标区域锁定范围小且准确以及等待时长缩短的目标，进一步提高目标车辆从目标车辆所在的区域变道至目标区域的效率。
在本公开的一些实施例中，为了进一步提高目标车辆从目标车辆所在的区域变道至目标区域的效率，如图8所示，最优解求解模型的训练方法包括步骤S801至步骤S805。
S801：获取所有车道内目标车辆对应的交通事故概率值、车辆密集程度和目标区域的初始化参数。
S802：根据初始化参数，获得初始化参数矩阵。
S803：基于麻雀搜索算法，根据第二交通事故发生概率阈值，更新初始化参数矩阵。
S804：将更新后的初始化参数矩阵代入机器学习算法计算，当更新后的初始化参数矩阵满足多目标函数优化模型，输出调整后的所有车道内目标车辆对应的目标区域对应的加权参数。
S805：根据调整后的所有车道内车辆对应的目标区域对应的参数，训练最优解求解模型。
多目标函数优化模型表示如下：

其中，f.表示多目标函数优化模型中的多个目标函数；表示锁定范围最小且准确的目标区域；表示最短等待时长；表示最低交通事故概率；表示最高车辆密集程度；s.t.表示多目标函数优化模型中的多个约束条件。
在一个实施例中，最优解求解模型可以是基于麻雀搜索算法改进的卷积神经网络算法。最优解求解模型也可以基于其他的人工智能算法和机器学习算法，在此不做限定。
为便于理解，在对步骤S801至步骤S805进行说明之前，先对麻雀搜索算法改进的卷积神经网络算法进行说明。
在麻雀搜索算法中，在麻雀搜索食物的过程中，麻雀种群分为发现者和加入者，发现者在种群中负责寻找食物并为整个麻雀种群提供觅食区域和方向，而加入者则是利用发现者来获取食物，当麻雀种群意识到危险时会做出反捕食行为。
在卷积神经网络算法中，卷积神经网络是一种使用卷积计算且具有深度结构的前馈神经网络，能够处理复杂的时空序列数据，以前馈结果迭代指导下一次隐含层输入并进行结果预测。
基于麻雀搜索改进的卷积神经网络算法，结合了麻雀搜索算法和卷积神经网络算法的优点，能够以历史数据的目标车辆作为麻雀发现者，为后续目标车辆作为加入者提供指导，如果遇到交通事故概率大的危险场景以反捕食行为进行避免，卷积神经网络通过麻雀搜索算法进行深度学习，经过卷积层、池化层和全连接层的迭代运算输出最优的数值。
接下来，将以基于麻雀搜索算法改进的卷积神经网络算法，对公式(8)所示的多目标函数优化模型求最优解，对最优解求解模型的训练过程进行说 明。
假设，开始时第i个车道的目标车辆和交通区域情况输入初始化参数，用如下向量：
那么所有车道的目标车辆和交通区域情况输入初始化参数，可以用如下矩阵表示：
随后，对矩阵Y进行归一化处理。
基于麻雀搜索算法，矩阵Y中探索者(即发现者)的位置迭代更新方式如下：
其中，表示在t时刻，麻雀个体(即目标车辆)的位置；i表示当前迭代次数，itermax表示最大迭代次数；δ∈[0,1]内的一个随机数；Pθ表示预警安全阈值(即第二交通事故发生概率阈值)；ρ表示服从正态分布的随机数；Γ表示每个元素均为1的矩阵。
基于麻雀搜索算法，矩阵Y中追随者(即加入者)的位置迭代更新方式如下：
其中，表示在t时刻，当前全局最差探索者的位置；n表示种群规模；表示在t+1时刻，当前全局最优探索者的位置，A+表示每个元素随机幅值1或-1的矩阵。
将上述矩阵Y代入卷积神经网络的卷积层，按如下公式计算处理：
其中，Zl和Zl+1表示第l+1层的卷积输入和输出，即特征图；K为特征图的通道数，f、s0和b表示卷积层参数和偏差量。
然后，在卷积层进行特征提取后，输出的特征图会被传递至池化层进行特征选择和信息过滤。
池化层模型将特征图中单个点的结果替换为其相邻区域的特征图统计量， 计算公式如下：
其中，μ表示预指定参数取池化区域内的均值或极大值。
然后，全连接层对卷积层提取的特征进行非线性组合以得到输出，最后输出层输出包括γ、α、β、ci等数值。
如果最后输出层输出的包括γ、α、β和ci等数值，对应达到自动驾驶发生交通事故的概率低、车辆密集程度高、锁定的目标区域小和等待时间短为目标，即若最后输出层输出的包括γ、α、β和ci等数值满足公式(8)，则接受最后输出层输出的包括γ、α、β和ci等数值为最优数值，否则继续通过基于麻雀搜索改进的卷积神经网络算法进行迭代求解。
基于麻雀搜索改进的卷积神经网络算法，输出包括γ、α、β和ci等满足多目标函数优化模型的数值，达到自动驾驶发生交通事故的概率低、车辆密集程度高、锁定的目标区域小和等待时间短的目标，能够进一步提高目标车辆从目标车辆所在的区域变道至目标区域的效率。
在本公开的一些实施例中，为了进一步提高目标车辆从目标车辆所在的区域变道至目标区域的效率，本公开实施例提供的交通区域锁控制方法还可以包括步骤S901。
S901：获取并发送目标车辆的变道失败率。
MEC节点根据历史数据，获取目标车辆的变道失败率，并将变道失败率自动传输给道路交通运输管理部门，交由道路交通运输管理部门进行大数据分析，以便于道路交通运输管理部门考虑采用道路过度拥堵需要进行合理引流，还是道路太窄应该增加车道等适当的改进措施，来解决交通问题。
根据历史数据，基于下式，获取目标车辆的变道失败率：
其中，Si表示目标车辆的变道失败率；Sf表示目标车辆变道失败；Ss表示目标车辆变道成功。
在一个实施例中，基于公式(15)，还可以计算目标区域在任意时间范围内所有目标车辆变道失败率。
通过获取并发送目标车辆的变道失败率，使得道路交通运输管理部门采用适当的改进措施，来解决交通问题，进一步提高目标车辆从目标车辆所在 的区域变道至目标区域的效率。
基于上述实施例提供的交通区域锁控制方法，相应地，本公开还提供了交通区域锁控制装置的具体实现方式。请参见以下实施例。
首先参见图9，本公开实施例提供的交通区域锁控制装置包括：
获取模块901，用于获取目标车辆的多维时空位置；
确定模块902，用于根据多维时空位置，确定目标车辆所在的区域；
确定模块902，还用于根据目标车辆所在的区域的投影区域和预设偏移量，确定偏移区域；
确定模块902，还用于根据投影区域和偏移区域，确定目标区域；
控制模块903，用于当目标车辆所在的区域处于目标区域的邻区，且目标区域的区域锁为空闲状态时，控制目标区域内的区域锁为占用状态，并控制目标车辆从目标车辆所在的区域变道至目标区域。
通过获取目标车辆的多维时空位置；再根据多维时空位置，确定目标车辆所在的区域；再根据目标车辆所在的区域的投影区域和预设偏移量，确定偏移区域；再根据投影区域和偏移区域，确定目标区域；当目标车辆所在的区域处于目标区域的邻区，且目标区域的区域锁为空闲状态时，控制目标区域内的区域锁为占用状态，并控制目标车辆从目标车辆所在的区域变道至目标区域。如此，根据目标车辆的多维时空位置，确定目标车辆的目标区域，控制目标区域内的区域锁从空闲状态变更为占用状态，并控制目标车辆从目标车辆所在的区域变道至目标区域，使得该目标区域仅供目标车辆变道，能够避免多个车辆从多个不同的车道抢占同一车道相同位置导致的交通事故的情况。
在本公开的一些实施例中，为了进一步精确目标车辆所在的区域，以便提高控制目标车辆从目标车辆所在的区域变道至目标区域的效率，如图10所示，确定模块902可以包括：
获取单元，用于根据多维时空位置，获取目标车辆多个预设方向的投影点位；
确定单元，用于根据投影点位，确定所述区域。
在本公开的一些实施例中，通过划分网格或GPS等定位方法，也可以获得区域的投影区域，但会出现因网格过大不准确、GPS信号差或者距离远误 差大的问题。为了进一步精确偏移区域，以便提高控制目标车辆从目标车辆所在的区域变道至目标区域的效率，如图10所示，确定模块902中确定单元还可以用于：
根据目标车辆所在的区域的投影的位置关系，确定投影区域；
根据预设偏移量和投影区域，确定偏移区域。
在本公开的一些实施例中，为了进一步避免多个车辆从多个不同的车道抢占同一车道相同位置导致的交通事故的情况，当目标区域内的区域锁为占用状态时，如图10所示，本公开实施例提供的交通区域锁控制装置还可以包括：
接收模块1001，用于接收目标车辆的区域锁请求，区域锁请求为目标车辆基于目标车辆的车辆信息和目标区域发送的区域锁请求；
添加模块1002，用于将区域锁请求添加至目标区域的区域锁请求列表。
控制模块903，还用于当检测到目标区域的区域锁为空闲状态，且按照区域锁请求列表中的顺序达到目标车辆的区域锁请求时，控制目标车辆从目标车辆所在的区域变道至目标区域。
在本公开的一些实施例中，为了优先处理110警车、120医疗救护车、119消防救火车等救援车辆进行变道，区域锁请求还包括目标车辆的车辆类型，如图10所示，确定模块902，还用于当接收到目标车辆的区域锁请求时，基于预设的车辆类型与优先级的对应关系，确定目标车辆的优先级。添加模块1002，还用于基于优先级将区域锁请求添加至区域锁请求列表。
在本公开的一些实施例中，为了便于用户了解变道所需要的等待的时长，丰富MEC节点所呈现的功能，如图10所示，获取模块，还用于获取第一请求时间，第一请求时间为区域锁请求添加至区域锁请求列表的请求时间。获取模块，还用于获取第二请求时间，第二请求时间为当前占用目标区域的车辆的区域锁请求添加至区域锁请求列表的请求时间。
本公开实施例提供的交通区域锁控制装置还可以包括：获得模块1003，用于计算第一请求时间和第二请求时间的差值，获得目标车辆的等待时长。
控制模块903，还用于控制目标车辆在等待时长后，从目标车辆所在的区域变道至目标区域。
在本公开的一些实施例中，为了提高目标车辆从目标车辆所在的区域变 道至目标区域的效率，如图10所示，获得模块1003还用于：
将交通事故概率值、车辆密集程度和目标区域输入最优解求解模型，获得调整后的目标区域对应的加权参数；
根据调整后的目标区域对应的加权参数，调整目标区域，获得调整后的目标区域。
在本公开的一些实施例中，为了进一步提高目标车辆从目标车辆所在的区域变道至目标区域的效率，如图10所示，控制模块903，还用于：
当目标车辆所在的区域处于调整后的目标区域的邻区，且调整后的目标区域的区域锁为空闲状态时，控制调整后的目标区域内的区域锁为占用状态，并控制目标车辆从目标车辆所在的区域变道至调整后的目标区域。
在本公开的一些实施例中，为了进一步提高目标车辆从目标车辆所在的区域变道至目标区域的效率，如图10所示，获得单元1003包括：
获取单元，用于获取所有车道内目标车辆对应的交通事故概率值、车辆密集程度和目标区域的初始化参数；
获得单元，用于根据初始化参数，获得初始化参数矩阵；
更新单元，用于基于麻雀搜索算法，根据第二交通事故发生概率阈值，更新初始化参数矩阵；
输出单元，用于将更新后的初始化参数矩阵代入机器学习算法计算，当更新后的初始化参数矩阵满足多目标函数优化模型，输出调整后的所有车道内目标车辆对应的目标区域对应的加权参数；
训练单元，用于根据调整后的所有车道内车辆对应的目标区域对应的参数，训练最优解求解模型。
图11示出了本公开实施例提供的交通区域锁控制的硬件结构示意图。
在交通区域锁控制设备可以包括处理器1101以及存储有计算机程序指令的存储器1102。
具体地，上述处理器1101可以包括中央处理器(CPU)，或者特定集成电路(Application Specific Integrated Circuit，ASIC)，或者可以被配置成实施本公开实施例的一个或多个集成电路。
存储器1102可以包括用于数据或指令的大容量存储器。举例来说而非限制，存储器1102可包括硬盘驱动器(Hard Disk Drive，HDD)、软盘驱动器、 闪存、光盘、磁光盘、磁带或通用串行总线(Universal Serial Bus，USB)驱动器或者两个或更多个以上这些的组合。在合适的情况下，存储器1102可包括可移除或不可移除(或固定)的介质。在合适的情况下，存储器1102可在交通区域锁控制设备的内部或外部。在特定实施例中，存储器1102是非易失性固态存储器。
在特定实施例中，存储器1102可包括只读存储器(ROM)，随机存取存储器(RAM)，磁盘存储介质设备，光存储介质设备，闪存设备，电气、光学或其他物理/有形的存储器存储设备。因此，通常，存储器1102包括一个或多个编码有包括计算机可执行指令的软件的有形(非暂态)计算机可读存储介质(例如，存储器设备)，并且当该软件被执行(例如，由一个或多个处理器)时，其可操作来执行参考根据本公开公开的第一方面的方法所描述的操作。
处理器1101通过读取并执行存储器1102中存储的计算机程序指令，以实现上述实施例中的任意一种交通区域锁控制方法。
在一个示例中，交通区域锁控制设备还可包括通信接口1103和总线1110。其中，如图11所示，处理器1101、存储器1102、通信接口1103通过总线1110连接并完成相互间的通信。
通信接口1103，主要用于实现本公开实施例中各模块、装置、单元和/或设备之间的通信。
总线1110包括硬件、软件或两者，将交通区域锁控制设备的部件彼此耦接在一起。举例来说而非限制，总线可包括加速图形端口(AGP)或其他图形总线、增强工业标准架构(EISA)总线、前端总线(FSB)、超传输(HT)互连、工业标准架构(ISA)总线、无限带宽互连、低引脚数(LPC)总线、存储器总线、微信道架构(MCA)总线、外围组件互连(PCI)总线、PCI-Express(PCI-X)总线、串行高级技术附件(SATA)总线、视频电子标准协会局部(VLB)总线或其他合适的总线或者两个或更多个以上这些的组合。在合适的情况下，总线1110可包括一个或多个总线。尽管本公开实施例描述和示出了特定的总线，但本公开考虑任何合适的总线或互连。
另外，结合上述实施例中的交通区域锁控制方法，本公开实施例可提供一种计算机存储介质来实现。该计算机存储介质上存储有计算机程序指令； 该计算机程序指令被处理器执行时实现上述实施例中的任意一种交通区域锁控制方法。
需要明确的是，本公开并不局限于上文所描述并在图中示出的特定配置和处理。为了简明起见，这里省略了对已知方法的详细描述。在上述实施例中，描述和示出了若干具体的步骤作为示例。但是，本公开的方法过程并不限于所描述和示出的具体步骤，本领域的技术人员可以在领会本公开的精神后，作出各种改变、修改和添加，或者改变步骤之间的顺序。
以上所述的结构框图中所示的功能块可以实现为硬件、软件、固件或者它们的组合。当以硬件方式实现时，其可以例如是电子电路、专用集成电路(ASIC)、适当的固件、插件、功能卡等等。当以软件方式实现时，本公开的元素是被用于执行所需任务的程序或者代码段。程序或者代码段可以存储在机器可读介质中，或者通过载波中携带的数据信号在传输介质或者通信链路上传送。“机器可读介质”可以包括能够存储或传输信息的任何介质。机器可读介质的例子包括电子电路、半导体存储器设备、ROM、闪存、可擦除ROM(EROM)、软盘、CD-ROM、光盘、硬盘、光纤介质、射频(RF)链路，等等。代码段可以经由诸如因特网、内联网等的计算机网络被下载。
还需要说明的是，本公开中提及的示例性实施例，基于一系列的步骤或者装置描述一些方法或系统。但是，本公开不局限于上述步骤的顺序，也就是说，可以按照实施例中提及的顺序执行步骤，也可以不同于实施例中的顺序，或者若干步骤同时执行。
上面参考根据本公开公开的实施例的方法、装置(系统)和计算机程序产品的流程图和/或框图描述了本公开公开的各方面。应当理解，流程图和/或框图中的每个方框以及流程图和/或框图中各方框的组合可以由计算机程序指令实现。这些计算机程序指令可被提供给通用计算机、专用计算机、或其它可编程数据处理装置的处理器，以产生一种机器，使得经由计算机或其它可编程数据处理装置的处理器执行的这些指令使能对流程图和/或框图的一个或多个方框中指定的功能/动作的实现。这种处理器可以是但不限于是通用处理器、专用处理器、特殊应用处理器或者现场可编程逻辑电路。还可理解，框图和/或流程图中的每个方框以及框图和/或流程图中的方框的组合，也可以由执行指定的功能或动作的专用硬件来实现，或可由专用硬件和计算机 指令的组合来实现。
以上所述，仅为本公开的具体实施方式，所属领域的技术人员可以清楚地了解到，为了描述的方便和简洁，上述描述的系统、模块和单元的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。应理解，本公开的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本公开揭露的技术范围内，可轻易想到各种等效的修改或替换，这些修改或替换都应涵盖在本公开的保护范围之内。

Claims (14)

  1. 一种交通区域锁控制方法，包括：
    获取目标车辆的多维时空位置；
    根据所述多维时空位置，确定所述目标车辆所在的区域；
    根据所述区域的投影区域和预设偏移量，确定偏移区域；
    根据所述投影区域和所述偏移区域，确定目标区域；
    当所述区域处于所述目标区域的邻区，且所述目标区域的区域锁为空闲状态时，控制所述目标区域内的区域锁为占用状态，并控制所述目标车辆从所述区域变道至所述目标区域。
  2. 根据权利要求1所述的方法，其中，在获取目标车辆的多维时空位置之前，所述方法还包括：
    获取目标车辆所在单位面积内的地理区域的车辆密集程度和交通事故发生概率；
    当所述车辆密集程度超过车辆密集程度阈值，和/或所述交通事故发生概率超过第一交通事故发生概率阈值时，获取目标车辆的多维时空位置。
  3. 根据权利要求2所述的方法，其中，当所述目标区域内的区域锁为占用状态时，所述方法还包括：
    接收所述目标车辆的区域锁请求，所述区域锁请求为所述目标车辆基于所述目标车辆的车辆信息和所述目标区域发送的区域锁请求；
    将所述区域锁请求添加至所述目标区域的区域锁请求列表；
    当检测到所述目标区域的区域锁为空闲状态，且按照所述区域锁请求列表中的顺序达到所述目标车辆的区域锁请求时，控制所述目标车辆从所述区域变道至所述目标区域。
  4. 根据权利要求3所述的方法，其中，所述区域锁请求包括所述目标车辆的车辆类型；所述方法还包括：
    当接收到所述目标车辆的区域锁请求时，基于预设的车辆类型与优先级的对应关系，确定所述目标车辆的优先级；
    基于所述优先级将所述区域锁请求添加至所述区域锁请求列表。
  5. 根据权利要求4所述的方法，其中，在所述将所述区域锁请求添加至 所述目标区域的区域锁请求列表之后，所述方法还包括：
    获取第一请求时间，所述第一请求时间为所述区域锁请求添加至所述区域锁请求列表的请求时间；
    获取第二请求时间，所述第二请求时间为当前占用所述目标区域的车辆的区域锁请求添加至所述区域锁请求列表的请求时间；
    计算所述第一请求时间和所述第二请求时间的差值，获得所述目标车辆的等待时长；
    控制所述目标车辆在所述等待时长后，从所述区域变道至所述目标区域。
  6. 根据权利要求5所述的方法，其中，在所述根据所述投影区域和所述偏移区域，确定目标区域之后，所述方法还包括：
    将所述交通事故概率值、所述车辆密集程度和所述目标区域输入最优解求解模型，获得调整后的目标区域对应的加权参数；
    根据所述调整后的目标区域对应的加权参数，调整所述目标区域，获得调整后的目标区域。
  7. 根据权利要求6所述的方法，其中，所述当所述区域处于所述目标区域的邻区，且所述目标区域的区域锁为空闲状态时，控制所述目标区域内的区域锁为占用状态，并控制所述目标车辆从所述区域变道至所述目标区域，包括：
    当所述区域处于所述调整后的目标区域的邻区，且所述调整后的目标区域的区域锁为空闲状态时，控制所述调整后的目标区域内的区域锁为占用状态，并控制所述目标车辆从所述区域变道至所述调整后的目标区域。
  8. 根据权利要求6或7所述的方法，其中，所述最优解求解模型的训练方法，包括：
    获取所有车道内目标车辆对应的交通事故概率值、车辆密集程度和目标区域的初始化参数；
    根据所述初始化参数，获得初始化参数矩阵；
    基于麻雀搜索算法，根据第二交通事故发生概率阈值，更新所述初始化参数矩阵；
    将更新后的所述初始化参数矩阵代入机器学习算法计算，当更新后的所述初始化参数矩阵满足多目标函数优化模型，输出调整后的所有车道内目标 车辆对应的目标区域对应的加权参数；
    根据调整后的所有车道内车辆对应的目标区域对应的参数，训练所述最优解求解模型。
  9. 根据权利要求1所述的方法，其中，所述根据所述多维时空位置，确定所述目标车辆所在的区域，包括：
    根据所述多维时空位置，获取所述目标车辆多个预设方向的投影点位；
    根据所述投影点位，确定所述区域。
  10. 根据权利要求1所述的方法，其中，所述根据所述区域的投影区域和预设偏移量，确定偏移区域，包括：
    根据所述区域的投影的位置关系，确定所述投影区域；
    根据所述预设偏移量和所述投影区域，确定偏移区域。
  11. 一种交通区域锁控制装置，所述装置包括：
    获取模块，用于获取目标车辆的多维时空位置，；
    确定模块，用于根据所述多维时空位置，确定所述目标车辆所在的区域；
    确定模块，还用于根据所述区域的投影区域和预设偏移量，确定偏移区域；
    确定模块，还用于根据所述投影区域和所述偏移区域，确定目标区域；
    控制模块，用于当所述区域处于所述目标区域的邻区，且所述目标区域的区域锁为空闲状态时，控制所述目标区域内的区域锁为占用状态，并控制所述目标车辆从所述区域变道至所述目标区域。
  12. 一种交通区域锁控制设备，所述设备包括：处理器以及存储有计算机程序指令的存储器；
    所述处理器执行所述计算机程序指令时实现如权利要求1-10任意一项所述的方法。
  13. 一种计算机可读存储介质，所述计算机可读存储介质上存储有计算机程序指令，所述计算机程序指令被处理器执行时实现如权利要求1-10任意一项所述的方法。
  14. 一种计算机程序产品，所述计算机程序产品中的指令由电子设备的处理器执行时，使得所述电子设备执行如权利要求1-10任意一项所述的方法。
PCT/CN2023/111650 2022-08-17 2023-08-08 交通区域锁控制方法、装置、设备和存储介质 Ceased WO2024037375A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210987081.3 2022-08-17
CN202210987081.3A CN117636684A (zh) 2022-08-17 2022-08-17 交通区域锁控制方法、装置、设备和存储介质

Publications (1)

Publication Number Publication Date
WO2024037375A1 true WO2024037375A1 (zh) 2024-02-22

Family

ID=89940634

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/111650 Ceased WO2024037375A1 (zh) 2022-08-17 2023-08-08 交通区域锁控制方法、装置、设备和存储介质

Country Status (2)

Country Link
CN (1) CN117636684A (zh)
WO (1) WO2024037375A1 (zh)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109895776A (zh) * 2017-12-11 2019-06-18 福特全球技术公司 车辆车道变换
CN110103967A (zh) * 2019-05-15 2019-08-09 广州小鹏汽车科技有限公司 一种车辆自动变道方法及车辆控制系统、车辆
CN112706763A (zh) * 2019-10-08 2021-04-27 现代自动车株式会社 车辆及控制该车辆的方法
CN113911115A (zh) * 2020-07-10 2022-01-11 丰田自动车工程及制造北美公司 自主车辆、系统和操作自主车辆的方法
CN114179832A (zh) * 2021-12-29 2022-03-15 阿波罗智联(北京)科技有限公司 用于自动驾驶车辆的变道方法

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102016203662A1 (de) * 2016-03-07 2017-09-07 Bayerische Motoren Werke Aktiengesellschaft Fahrerassistenzsystem und Verfahren zum Betreiben eines solchen
KR102368604B1 (ko) * 2017-07-03 2022-03-02 현대자동차주식회사 Ecu, 상기 ecu를 포함하는 무인 자율 주행 차량, 및 이의 차선 변경 제어 방법
CN107844117B (zh) * 2017-10-23 2020-10-30 上海木木聚枞机器人科技有限公司 一种基于云端的道路锁定系统及方法
DE102018132523A1 (de) * 2018-12-17 2020-06-18 Trw Automotive Gmbh Verfahren sowie System zum Steuern eines Kraftfahrzeugs
CN112416004B (zh) * 2020-11-19 2021-12-14 腾讯科技（深圳）有限公司 一种基于自动驾驶的控制方法、装置、车辆以及相关设备

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109895776A (zh) * 2017-12-11 2019-06-18 福特全球技术公司 车辆车道变换
CN110103967A (zh) * 2019-05-15 2019-08-09 广州小鹏汽车科技有限公司 一种车辆自动变道方法及车辆控制系统、车辆
CN112706763A (zh) * 2019-10-08 2021-04-27 现代自动车株式会社 车辆及控制该车辆的方法
CN113911115A (zh) * 2020-07-10 2022-01-11 丰田自动车工程及制造北美公司 自主车辆、系统和操作自主车辆的方法
CN114179832A (zh) * 2021-12-29 2022-03-15 阿波罗智联(北京)科技有限公司 用于自动驾驶车辆的变道方法

Also Published As

Publication number Publication date
CN117636684A (zh) 2024-03-01

Similar Documents

Publication Publication Date Title
EP3900412B1 (en) Adaptive multi-network vehicle architecture
US11915593B2 (en) Systems and methods for machine learning based collision avoidance
US11895566B2 (en) Methods of operating a wireless data bus in vehicle platoons
EP4129789A1 (en) Method and device for recognizing driving behavior of vehicle
US10375618B2 (en) Wireless communication apparatus and wireless communication method
US20210188311A1 (en) Artificial intelligence mobility device control method and intelligent computing device controlling ai mobility
US20180088571A1 (en) Autonomous vehicle operations with automated assistance
KR102203475B1 (ko) 자율 주행 시스템에서 차량을 제어하기 위한 방법 및 장치
US10567917B2 (en) System and method for indicating drones flying overhead
KR102807639B1 (ko) 자율주행시스템에서 지도정보를 이용하여 객체추적을 하는 방법 및 이를 위한 장치
KR20190107277A (ko) 자율 주행 시스템에서 차량을 제어하는 방법 및 장치
US10524187B2 (en) Wireless communication apparatus and wireless communication method using wireless LAN client
JP2019537159A5 (zh)
KR102606258B1 (ko) 협력 주행 제어 장치 및 방법
US11410561B2 (en) Traffic management systems and methods for unmanned aerial vehicles
CN112583872B (zh) 一种通信方法及装置
KR102836057B1 (ko) 응급차량 이동 경로 확보 방법 및 이를 구현한 시스템
CN108512880A (zh) 通信方法和通信设备
CN113643534B (zh) 交通管控的方法及设备
US20200029191A1 (en) Method and apparatus for setting a server bridge in an autonomous driving system
US20210326703A1 (en) On-board unit, method for cooperative driving, model determination unit, method for determining a machine-learning communication model, system, method, vehicle, and user equipment
CN113905941B (zh) 地理控制区中的车辆控制
US20240355202A1 (en) Vehicle travel control system, server device used thereby, and vehicle
US20250252850A1 (en) Misbehavior detection in a vehicle-to-everything (v2x) maneuver coordination message
US20230258763A1 (en) Radio frequency sensing to detect wireless devices and user presence by an autonomous vehicle

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: 23854278

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 23854278

Country of ref document: EP

Kind code of ref document: A1