WO2022068558A1 - 一种地图数据的传输方法及装置 - Google Patents

一种地图数据的传输方法及装置 Download PDF

Info

Publication number
WO2022068558A1
WO2022068558A1 PCT/CN2021/117860 CN2021117860W WO2022068558A1 WO 2022068558 A1 WO2022068558 A1 WO 2022068558A1 CN 2021117860 W CN2021117860 W CN 2021117860W WO 2022068558 A1 WO2022068558 A1 WO 2022068558A1
Authority
WO
WIPO (PCT)
Prior art keywords
probability
target
prior probability
target information
prior
Prior art date
Application number
PCT/CN2021/117860
Other languages
English (en)
French (fr)
Inventor
伍勇
刘建琴
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Publication of WO2022068558A1 publication Critical patent/WO2022068558A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/29—Geographical information databases
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/02—Protocols based on web technology, e.g. hypertext transfer protocol [HTTP]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network

Definitions

  • the present application relates to the technical field of data transmission, and in particular, to a method and device for transmitting map data.
  • Self-driving cars rely on the synergy of technologies such as artificial intelligence, visual computing, radar, positioning systems, and high-precision maps to allow computers to operate motor vehicles autonomously and safely without any human initiative.
  • technologies such as artificial intelligence, visual computing, radar, positioning systems, and high-precision maps to allow computers to operate motor vehicles autonomously and safely without any human initiative.
  • the accuracy and precision of high-precision maps and the efficiency of updates are critical to the safety of autonomous vehicles. For example, when roads are closed, routes change, or traffic signs change, maps need to be updated.
  • crowdsourcing has been widely adopted as a low-cost data collection mode in recent years.
  • Crowdsourcing is to complete a certain task based on the power of the public, for example, to complete the collection of map data based on a large number of Volkswagen vehicles.
  • the vehicle can report the detection information of each sensor to the in-vehicle fusion unit, but the reported information is in a sensor-specific form, that is, different sensors have different reporting contents and formats.
  • the vehicle reports the target result information detected by the sensor to the cloud. Note that the detection result information here only supports one type of target, where the target is an object in the natural environment.
  • each vehicle reports the detected target information.
  • confidence information is generally also reported to judge the credibility of the target information.
  • the cloud After the cloud gets the confidence value, it can fuse the reported results of multiple vehicles.
  • the most commonly used fusion methods in the industry include methods such as Bayesian decision-making based on the minimum error rate.
  • the cloud can be based on multiple vehicles.
  • the a posteriori probability of the target to be detected is calculated and the probability value is taken as the final confidence value.
  • the defects of Bayesian decision-making in the above technology include: as shown in Figure 1, the dependence of the accuracy of the prior hypothesis is too high. When the prior hypothesis is correct, the confidence curve does not exceed the upper limit of credibility, and the confidence curve is credible. However, when the prior hypothesis is incorrect, the confidence curve exceeds the upper limit of credibility, and the confidence curve cannot correctly reflect the relationship between the confidence and the error rate, that is, the estimation of the confidence may be invalid. The judgment of the existence of the target produces an error.
  • the embodiments of the present application provide a method and device for updating map data, which can minimize the error rate of cloud data fusion and improve the accuracy of data fusion.
  • a data transmission method comprising: determining at least one prior probability of a map element corresponding to target information, wherein the target information is information about the target detected by the vehicle sensing the surrounding environment , the prior probability is the probability that the target exists under the first condition, and each prior probability in the at least one prior probability corresponds to one of the first conditions;
  • the target information, the at least one prior probability, and the first condition corresponding to each prior probability are transmitted.
  • the first condition is the type of the map element corresponding to the target information, the location of the map element, the detection time of the map element, the sensor type, or one of the design operating conditions at least one.
  • the sensor type includes at least one of a lidar, a camera, a millimeter-wave radar, or an ultrasonic wave.
  • the corresponding relationship includes a table or a formula.
  • the first conditions corresponding to different sensor types are different.
  • the first condition when the sensor type is a camera, the first condition includes a lighting condition corresponding to the target information, a longitudinal distance corresponding to the target information, and the target information at least one of the corresponding lateral distance or the weather information corresponding to the target information.
  • the first condition corresponding to the prior probability includes the reflectivity corresponding to the target information, the longitudinal direction corresponding to the target information At least one of the distance, the lateral distance corresponding to the target information, or the weather information corresponding to the target information.
  • the first condition corresponding to the prior probability includes the interference level corresponding to the target information, the interference level corresponding to the target information At least one of a vertical distance and a horizontal distance corresponding to the target information.
  • the prior probability is a quantized value of the prior probability in the Bayesian estimation algorithm.
  • a data transmission device comprising:
  • the determination module is used to determine at least one prior probability of the map element corresponding to the target information, wherein the target information is the information about the target detected by the vehicle sensing the surrounding environment, and the prior probability is that the target is in the first A probability that exists under a condition, each of the at least one prior probability corresponds to one of the first conditions;
  • a sending module configured to send the target information, the at least one prior probability and the first condition corresponding to each prior probability.
  • the first condition includes the type of the map element corresponding to the target information, the position of the map element, the detection time of the map element, the sensor type, or any of the design operating conditions at least one.
  • the sensor type includes at least one of a lidar, a camera, a millimeter-wave radar, or an ultrasonic wave.
  • the corresponding relationship includes a table or a formula.
  • the first conditions corresponding to different sensor types are different.
  • the first condition includes a lighting condition corresponding to the target information, a longitudinal distance corresponding to the target information, and the target information at least one of the corresponding lateral distance or the weather information corresponding to the target information.
  • the first condition corresponding to the prior probability includes the reflectivity corresponding to the target information, the longitudinal direction corresponding to the target information At least one of the distance, the lateral distance corresponding to the target information, or the weather information corresponding to the target information.
  • the first condition corresponding to the prior probability includes the interference level corresponding to the target information, the interference level corresponding to the target information At least one of a vertical distance and a horizontal distance corresponding to the target information.
  • the prior probability is a quantized value of the prior probability in the Bayesian estimation algorithm.
  • the application provides a data processing method, the method comprising:
  • the receiving at least one prior probability includes:
  • the first prior probability is a prior probability in the state w1
  • the w1 characterizes the existence of the target
  • the second prior probability is w2
  • the prior probability of the state, the w2 characterizes the non-existence of the target.
  • the calculating a posteriori probability of the target information based on the at least one priori probability includes:
  • the first posterior probability is the posterior probability in the w1 state
  • the second posterior probability is the posterior probability in the w2 state.
  • the determining whether the target exists in the natural environment based on the posterior probability includes:
  • the first posterior probability is greater than the second posterior probability, it is determined that the target exists in the natural environment.
  • the first posterior probability is not greater than the second posterior probability, it is determined that the target does not exist in the natural environment.
  • the prior probability is a quantized value of the prior probability in the Bayesian estimation algorithm
  • the posterior probability is a quantized value of the posterior probability in the Bayesian estimation algorithm.
  • the Bayesian estimation based on the minimum error rate is performed on the vehicle report result, and the posterior probability in the state where the target exists is compared with the posterior probability in the state where the target does not exist, and then the decision is determined. It can minimize the error rate of cloud data fusion and improve the accuracy of data fusion.
  • the present application provides a data processing device, the device comprising:
  • a receiving module configured to receive target information, at least one prior probability and a first condition corresponding to each of the at least one prior probability, wherein the target information reflects the detection of the surrounding environment detected by the vehicle
  • the obtained information about the target, the prior probability is the probability that the target exists under the first condition, and each prior probability in the at least one prior probability corresponds to one of the first conditions;
  • a calculation module configured to calculate a posteriori probability of the target information based on the at least one priori probability
  • a determination module configured to determine whether the target exists in the natural environment based on the posterior probability.
  • the receiving module is used for:
  • the first prior probability is a prior probability in the state w1
  • the w1 characterizes the existence of the target
  • the second prior probability is w2
  • the prior probability of the state, the w2 characterizes the non-existence of the target.
  • the computing module is specifically used for:
  • the first posterior probability is the posterior probability in the w1 state
  • the second posterior probability is the posterior probability in the w2 state.
  • the determining module is specifically used for:
  • the first posterior probability is greater than the second posterior probability, it is determined that the target exists in the natural environment.
  • the first posterior probability is not greater than the second posterior probability, it is determined that the target does not exist in the natural environment.
  • the prior probability is a quantized value of the prior probability in the Bayesian estimation algorithm
  • the posterior probability is a quantized value of the posterior probability in the Bayesian estimation algorithm.
  • a data transmission device comprising: a memory for storing program instructions;
  • the processor is configured to execute the first aspect or any possible data transmission method of the first aspect when the program instructions in the memory are invoked and executed.
  • a readable storage medium comprising: an execution instruction is stored in the readable storage medium, and when at least one processor of the data transmission apparatus executes the execution instruction, the data transmission apparatus executes the first execution instruction.
  • an execution instruction is stored in the readable storage medium, and when at least one processor of the data transmission apparatus executes the execution instruction, the data transmission apparatus executes the first execution instruction.
  • a data processing device comprising: a memory for storing program instructions;
  • the processor is configured to execute the third aspect or any possible data processing method of the third aspect when the program instructions in the memory are invoked and executed.
  • a readable storage medium comprising: an execution instruction is stored in the readable storage medium, and when at least one processor of the data transmission apparatus executes the execution instruction, the data transmission apparatus executes the first execution instruction.
  • a server including the fourth aspect or any possible data processing apparatus of the fourth aspect.
  • a tenth aspect provides a chip, comprising at least one processor, the processor is coupled to a memory, the processor is configured to read instructions in the memory and execute any one of the first aspect or the third aspect according to the instructions method described in item.
  • a computer program product which, when run on a computer, causes the computer to perform the method described in any one of the first aspect or the fourth aspect.
  • a twelfth aspect provides a roadside unit, comprising any of the possible devices of the second aspect or the fourth aspect.
  • any of the data processing devices, readable storage media, computer program products, servers, chips, and roadside units provided above can be implemented by the corresponding methods provided above.
  • beneficial effects achieved reference may be made to the beneficial effects in the corresponding methods provided above, which will not be repeated here.
  • Fig. 1 is a Bayesian decision curve fused according to confidence in the prior art
  • FIG. 2 is a schematic diagram of an application scenario of a map data processing method provided by an embodiment of the present application
  • FIG. 3 is a schematic diagram of an application scenario of another map data processing method provided by an embodiment of the present application.
  • FIG. 4 is a schematic diagram of an application scenario of still another map data processing method provided by an embodiment of the present application.
  • FIG. 5 is a schematic diagram of the update structure of the command-side map data provided by the embodiment of the present application.
  • FIG. 6 is a schematic diagram of a vehicle-end observation target provided by an embodiment of the present application.
  • FIG. 7 is a flowchart of a data transmission method provided by an embodiment of the present application.
  • FIG. 8 is a schematic structural diagram of a data transmission module provided by an embodiment of the present application.
  • FIG. 9 is a flowchart of a data processing method provided by an embodiment of the present application.
  • FIG. 10 is a schematic structural diagram of a data processing apparatus according to an embodiment of the present application.
  • FIG. 11 provides a method for updating map data according to an embodiment of the present application.
  • FIG. 13 is a schematic structural diagram of a computer program product provided by an embodiment of the present application.
  • first and second are only used for descriptive purposes, and should not be construed as indicating or implying relative importance or implicitly indicating the number of indicated technical features.
  • a feature defined as “first” or “second” may expressly or implicitly include one or more of that feature.
  • plural means two or more.
  • words such as “exemplary” or “for example” are used to represent examples, illustrations or illustrations. Any embodiments or designs described in the embodiments of the present application as “exemplary” or “such as” should not be construed as preferred or advantageous over other embodiments or designs. Rather, the use of words such as “exemplary” or “such as” is intended to present the related concepts in a specific manner to facilitate understanding.
  • the present application provides a data transmission method, in which at least one prior probability of a map element corresponding to target information is determined at a vehicle terminal, wherein the target information is information about the target detected by the vehicle sensing the surrounding environment, and the priori
  • the test probability is the probability that the target exists under the first condition, and each prior probability in the at least one prior probability corresponds to one of the first conditions; the vehicle terminal sends the target information, the at least one prior probability and a first condition corresponding to each of the prior probability.
  • the present application also provides a method and device for processing map data, wherein the cloud server receives target information, at least one prior probability, and a first condition corresponding to each prior probability in the at least one prior probability from the vehicle terminal,
  • the target information is information about the target detected by the vehicle sensing the surrounding environment, the prior probability is the probability that the target exists under the first condition, and each of the at least one prior probability
  • a priori probability corresponds to one of the first conditions; a posteriori probability of the target information is calculated based on the at least one priori probability; based on the posteriori probability, it is determined whether the target exists in the natural environment, and if so The high-resolution map will be updated.
  • the cloud server 11 performs the Bayesian estimation based on the minimum error rate on the vehicle reporting results from different sources, compares the posterior probability in the state where the target exists and the posterior probability in the state where the target does not exist, and then determines the decision, which can maximize the Reduce the error rate of cloud data fusion to a certain extent and improve the accuracy of data fusion.
  • the map data processing method, device and system provided by the present application can be applied to the fields of unmanned driving, driver assistance (ADAS) or intelligent driving.
  • ADAS driver assistance
  • the map in this application can be an electronic map, which is a digital map, including a high-precision map.
  • An electronic map is a map based on a map database, using computer technology, stored in digital form, and can be displayed on the screen of a terminal device.
  • the main components of an electronic map are map elements, such as geographical elements such as mountains, water systems, land, administrative divisions, points of interest or roads, and other geographical elements such as lane lines, crosswalks, stop lines, traffic signs, road signs, light poles, traffic lights, gantry. , roundabout, parking lot and other target elements on roads, among which, roads can be further divided into five levels: expressways, first-class roads, second-class roads, third-class roads and fourth-class roads, and the roads of each class can be different map element.
  • the cloud server 11 can provide real-time map data for multiple vehicles 12 through a wireless network.
  • the cloud server 11 includes a large-capacity storage space for storing map data, including high-precision maps, and is responsible for updating and distributing electronic maps, etc. .
  • the map data can be deployed on one or more servers.
  • the network platform can decide whether to update the current map based on the road data reported by the crowdsourcing vehicle terminal 12 or the updated target map element, and perform the update work of the map data, and after the update, it can be updated. Issue a new electronic map.
  • the vehicle terminal 12 is a front-end device for vehicle communication and management, and can be installed in various vehicles. At least one vehicle terminal 12 may be a communication device in a large-scale social vehicle, or a communication device in an intelligent vehicle with data processing and identification capabilities, or may be composed of ordinary social vehicles and intelligent vehicles.
  • the crowdsourced vehicle terminal 12 can serve as an important basis for map collection and update, and can obtain real-time road detection information through on-board sensors configured in the vehicle terminal, for example, through cameras, infrared sensors, radar detectors, global satellite navigation systems or inertial The navigation system (inertial navigation for short), etc., can upload the detected road information to the cloud server 11 for real-time update and maintenance of map data. Specifically, the vehicle terminal 12 may upload the collected road data or updated target map elements to the cloud server 11 or other network devices through the network.
  • the communication system may further include a roadside unit 13, which communicates with the vehicle terminal 11 or the cloud server 12 through a wired or wireless network, and can provide the vehicle terminal 11 with road information within a certain area, high-precision Positioning, or providing services such as high-precision maps.
  • the roadside unit 13 can also be used to collect road detection information reported by vehicle terminals 11 within a certain area, and through data processing and fusion, determine whether the corresponding map elements have changed, so as to decide whether to update the map.
  • FIG. 2 is a schematic diagram of a scenario where the method for processing map data provided by the present application is applied.
  • the cloud server 11 and the vehicle terminal 12 are mainly involved.
  • the vehicle terminal 12 uses a local map and can perceive the surrounding environment.
  • the sensor provided on the vehicle terminal 12 can perceive the surrounding environment and detect target information.
  • the vehicle terminal 12 determines the prior probability according to the detected target information.
  • the cloud server 11 may store a high-precision map database, the cloud server 11 may receive at least two target information from the vehicle terminal 12, and the cloud is set to receive target information, at least one prior probability, and the at least one prior probability.
  • the posterior probability of the target information is calculated based on the at least one prior probability, and based on the posterior probability, it is determined whether the target exists in the natural environment, and further Determine whether to update the map.
  • FIG. 3 is a schematic diagram of another scenario where the method for processing map data provided by the present application is applied.
  • the cloud server 11, the vehicle terminal 12 and the base station or roadside unit device 13 are mainly involved.
  • the vehicle terminal 12 uses a local map and can perceive the surrounding environment.
  • the sensor provided on the vehicle terminal 12 can perceive the surrounding environment and detect target information.
  • the vehicle terminal 12 determines the prior probability according to the detected target information.
  • the cloud server 11 can store a high-precision map database, the cloud server 11 can receive at least two target information from the vehicle terminal 12, the cloud server 11 can receive at least two target information from the vehicle terminal 12, and the cloud is configured to receive target information, at least one prior probability and a first condition corresponding to each of the at least one prior probability, a posterior probability of the target information is calculated based on the at least one prior probability, based on the posterior probability probability, determine whether the target exists in the natural environment, and further determine whether to update the map. .
  • the cloud server 11 updates and delivers the high-precision map, and the roadside unit (RSU) 13 performs key perception on the unmatched target information according to the instructions issued by the cloud server 11, and reports the perception result.
  • RSU roadside unit
  • FIG. 4 is a schematic diagram of still another scenario where a method for updating map data provided by the present application is applied.
  • the vehicle terminal 12 and the base station or roadside unit device 13 are mainly involved, wherein the vehicle terminal 12 uses a local map and can perceive the surrounding environment.
  • the sensor provided on the vehicle terminal 12 can perceive the surrounding environment, detect target information, and the vehicle terminal 12 determines the prior probability according to the detected target information.
  • the vehicle terminal 12 sends the prior probability to the base station or the roadside unit device 13, and the roadside unit device 13 receives the target information, at least one prior probability, and a prior probability corresponding to each of the at least one prior probability
  • the first condition of the first condition is to calculate the posterior probability of the target information based on the at least one prior probability, and based on the posterior probability, determine whether the target exists in the natural environment, and further determine whether to update the map.
  • FIG. 5 is a schematic structural diagram of a cloud command-side map data update provided by an embodiment of the present application.
  • the on-board computer system 112 may also receive information from or transfer information to other computer systems.
  • sensor data collected from the sensor system of the vehicle terminal 12 may be transferred to another computer for processing of the data.
  • data from the computer system 112 may be transmitted via a network to a cloud-side computer 720 for further processing.
  • Networks and intermediate nodes may include various configurations and protocols, including the Internet, the World Wide Web, Intranets, Virtual Private Networks, Wide Area Networks, Local Area Networks, private networks using one or more of the company's proprietary communication protocols, Ethernet, WiFi and HTTP, and various combinations of the foregoing.
  • Such communications may be by any device capable of transferring data to and from other computers, such as modems and wireless interfaces.
  • computer 720 may include a server having multiple computers, such as a load balancing server farm, that exchange information with different nodes of the network for the purpose of receiving, processing, and transmitting data from computer system 112.
  • the server may be configured similarly to computer system 110 , with processor 730 , memory 740 , instructions 750 , and data 760 .
  • the data 760 may include the prior probability of the observed target, and the server 720 may receive, monitor, store, update, and various information related to the map data, and determine whether the map data is updated.
  • FIG. 7 is a data transmission method provided by an embodiment of the present application. As shown in FIG. 7 , the execution subject of this embodiment is a vehicle.
  • S101 Determine at least one prior probability of a map element corresponding to target information, where the target information is information about the target detected by the vehicle sensing the surrounding environment, and the prior probability is that the target is under a first condition a probability of existence, each of the at least one prior probability corresponds to one of the first conditions.
  • the target can be an object in the natural environment, specifically an object on the actual road that affects the driving of the vehicle terminal, such as the traffic lights beside the road, the buildings on both sides of the road, or the road signs, light poles, Traffic signs etc.
  • Map elements are elements that make up a map, and an electronic map includes multiple map elements.
  • the target information can be the information about the target detected by the vehicle terminal to perceive the surrounding environment.
  • the sensor installed on the vehicle terminal such as lidar, camera millimeter wave radar, ultrasonic wave or combined inertial navigation. Wait for the sensor to detect the surrounding environment to obtain target information.
  • the vehicle terminal 12 sends the detected target information to the cloud server 11 .
  • the prior probability is the probability that the target exists under the first condition
  • the first condition is the type of the map element corresponding to the target information, the location of the map element, the detection time of the map element, the sensor type corresponding to the target information, or At least one of the design operating condition (ODD) types corresponding to the target information.
  • ODD design operating condition
  • the vehicle end can report multiple prior probabilities, and the multiple prior probabilities have multiple different first conditions, such as the prior probability that the traffic lights beside the road exist at the intersection; in the case of intersection maintenance, it is assumed that the average three day to install traffic lights, prior probability of existence after three days, etc.
  • Each of the at least one prior probability corresponds to one of the first conditions.
  • the vehicle terminal needs to send three items of data to the cloud, including the target information, the prior probability to be reported, and the first condition corresponding to the prior probability. There may be one or more prior probabilities, and there may be one or more corresponding first conditions.
  • the data transmission can be carried out by using a self-defined data format, or using a general data format, or using a standard data format.
  • the reporting method can be transmitted by technologies such as network communication and V2X.
  • the network communication technology may include a wireless network, such as wifi, for transmission.
  • the first condition is at least one of the type of the map element corresponding to the target information, the location of the map element, the detection time of the map element, the type of sensor, or a design operating condition.
  • the sensor type includes at least one of lidar, camera, millimeter-wave radar, or ultrasonic.
  • the corresponding relationship includes a table or a formula, where the table may be shown in Tables 1-4 below.
  • the first conditions corresponding to different sensor types are different.
  • the first condition includes the lighting condition corresponding to the target information, the vertical distance corresponding to the target information, the horizontal distance corresponding to the target information, or the target information. At least one of the weather information corresponding to the information.
  • the first condition corresponding to the prior probability includes the reflectivity corresponding to the target information, the vertical distance corresponding to the target information, and the horizontal direction corresponding to the target information. At least one of distance or weather information corresponding to the target information.
  • the first condition corresponding to the prior probability includes the interference level corresponding to the target information, the longitudinal distance corresponding to the target information, and the At least one of the lateral distances.
  • the prior probability is a quantized value of the prior probability in the Bayesian estimation algorithm.
  • the prior probability of different conditions is used, a variety of influencing factors are comprehensively considered, and different factors affecting the existence of the target are considered for different target types, thereby improving the prior probability of the target information reported by the vehicle terminal. reliability.
  • the present application also provides a data sending apparatus 200, as shown in FIG. 8, which mainly includes the following modules:
  • the determining module 201 is configured to determine at least one prior probability of a map element corresponding to the target information, wherein the target information is information about the target detected by the vehicle sensing the surrounding environment, and the prior probability is that the target is at a probability that exists under the first condition, each of the at least one prior probability corresponds to one of the first conditions;
  • a sending module 202 configured to send the target information, the at least one prior probability, and the first condition corresponding to each prior probability.
  • the first condition is at least one of the type of the map element corresponding to the target information, the location of the map element, the detection time of the map element, the sensor type, or a design operating condition.
  • the sensor type includes at least one of lidar, camera, millimeter-wave radar, or ultrasonic.
  • the corresponding relationship includes a table or a formula, where the table may be shown in Tables 1-4 below.
  • the first conditions corresponding to different sensor types are different.
  • the first condition includes a lighting condition corresponding to the target information, a vertical distance corresponding to the target information, a horizontal distance corresponding to the target information, or the target information. At least one of weather information corresponding to the information.
  • the first condition corresponding to the prior probability includes the reflectivity corresponding to the target information, the vertical distance corresponding to the target information, and the horizontal direction corresponding to the target information. At least one of distance or weather information corresponding to the target information.
  • the first condition corresponding to the prior probability includes the interference level corresponding to the target information, the longitudinal distance corresponding to the target information, and the At least one of the lateral distances.
  • the prior probability is a quantized value of the prior probability in the Bayesian estimation algorithm.
  • FIG. 9 is a flowchart of a data processing method provided by this application.
  • the execution subject of this embodiment may be a cloud server 11 or a base station or a roadside unit device 13 , and the following takes the cloud server 11 as an example to perform describe.
  • the data processing method specifically includes the following steps:
  • S301 Receive target information, at least one prior probability, and a first condition corresponding to each of the at least one prior probability, wherein the target information is a target detected by the vehicle sensing the surrounding environment information, the prior probability is the probability that the target exists under the first condition, and each prior probability in the at least one prior probability corresponds to one of the first conditions.
  • the cloud server 11 may be a map cloud, and the specific action execution object may be a computing device in the cloud server, such as a processor.
  • the data received by the cloud server 11 is the data sent by the vehicle terminal according to a certain data format.
  • the data includes three items: target information, a priori probability of the target information, and a first condition corresponding to the prior probability.
  • the prior probabilities may be multiple prior probabilities based on different first conditions.
  • the format of the data received by the cloud server may adopt a custom data format, or a general data format, or a standard data format for transmission.
  • the reporting method can be transmitted by technologies such as network communication and V2X.
  • the network communication technology may include a wireless network, such as wifi, for transmission.
  • the prior probability is the probability that the event has not happened yet.
  • the posterior probability is the probability that something has happened, and the reason why it is required to happen is caused by a certain factor.
  • the posterior probability refers to the probability of re-correction after obtaining the information of the "result", which is the "effect” in the problem of "seeking the cause”.
  • the prior probability and the posterior probability are inseparably linked, and the calculation of the posterior probability should be based on the prior probability.
  • the calculation method of the posterior probability can be calculated by a mathematical model, for example, it can be calculated by a Bayesian formula.
  • the prior probability is a quantized value of the prior probability in the Bayesian estimation algorithm; or the posterior probability is the quantized value of the posterior probability in the Bayesian estimation algorithm.
  • a first posterior probability may be calculated based on the first prior probability; a second posterior probability may be calculated based on the second prior probability; wherein the first posterior probability is the posterior in the w1 state probability, the second posterior probability is the posterior probability in the w2 state.
  • the determination module of the cloud server 11 determines whether the target exists in the natural environment based on the result of the posterior probability.
  • the first posterior probability is greater than the second posterior probability, it is determined that the target exists in the natural environment.
  • the first posterior probability is not greater than the second posterior probability, it is determined that the target does not exist in the natural environment.
  • the Bayesian estimation based on the minimum error rate is performed on the reported results of vehicles from different sources, and the posterior probability in the state where the target exists is compared with the posterior probability in the state where the target does not exist, and then a decision is made. It can minimize the error rate of cloud data fusion and improve the accuracy of data fusion.
  • FIG. 10 is a schematic structural diagram of a data processing apparatus 400 provided by this application.
  • the data processing apparatus 400 mainly includes the following modules:
  • the receiving module 401 is configured to receive target information, at least one prior probability, and a first condition corresponding to each prior probability in the at least one prior probability, wherein the target information reflects a situation in which the vehicle perceives the surrounding environment.
  • the detected information about the target, the prior probability is the probability that the target exists under a first condition, and each prior probability in the at least one prior probability corresponds to one of the first conditions;
  • a calculation module 402 configured to calculate a posteriori probability of the target information based on the at least one priori probability
  • a determination module 403 configured to determine whether the target exists in the natural environment based on the posterior probability.
  • the receiving module is configured to: receive a first prior probability and a second prior probability, wherein the first prior probability is a prior probability in the w1 state, and the w1 represents the existence of the target , the second prior probability is the prior probability in the state of w2, and the w2 indicates that the target does not exist.
  • the calculation module is specifically configured to: calculate a first posterior probability based on the first prior probability; calculate a second posterior probability based on the second prior probability; wherein the first posterior probability is the The posterior probability in the w1 state, and the second posterior probability is the posterior probability in the w2 state.
  • the determining module is specifically configured to: when the first posterior probability is greater than the second posterior probability, determine that the target exists in the natural environment; or when the first posterior probability When not greater than the second posterior probability, it is determined that the target does not exist in the natural environment.
  • the prior probability is a quantized value of the prior probability in the Bayesian estimation algorithm; or the posterior probability is the quantized value of the posterior probability in the Bayesian estimation algorithm.
  • the present application also provides a method for updating map data.
  • the observations of each vehicle are assumed to be independent.
  • the method for updating map data in this embodiment is described below, which specifically includes the following steps:
  • the cloud server 11 determines the first initial probability and the second initial probability of the k-th target.
  • the cloud server 11 may be a map cloud, and the specific action execution object may be a computing device in the cloud server, such as a processor.
  • the target can be an object in the natural environment, specifically an object in the actual road that affects the driving of the vehicle terminal, such as traffic lights beside the road, buildings on both sides of the road, or road signs, light poles, traffic signs cards, etc.
  • Map elements are elements that make up a map, and an electronic map includes multiple map elements.
  • the target information can be the target information detected by the vehicle terminal to perceive the surrounding environment.
  • the sensors set on the vehicle terminal such as lidar, camera millimeter-wave radar, ultrasonic or combined inertial navigation Wait for the sensor to detect the surrounding environment to obtain target information.
  • the vehicle terminal 12 reports the detected target information to the cloud server 11, and the server 11 determines whether the target exists or not, and if so, adds the target to the map as a new map element.
  • the electronic map includes multiple elements, each element represents a target, the k-th target is one of the targets, and k is a positive integer.
  • the cloud server 11 receives the first prior probability and the second prior probability of the k-th target of the i-th vehicle among the N vehicles.
  • the vehicle terminals 12 participating in the crowdsourcing may include N vehicles, where N is a positive integer greater than or equal to 2, that is to say, at least two vehicles participate in the collection of target information.
  • the i-th vehicle among the N vehicles observes the k-th target through sensors, such as lidar, camera millimeter-wave radar, ultrasonic or combined inertial navigation, and finds the information of the target during the process of traveling or parking. It is inconsistent with the information in the map, for example, there is no element information representing the target in the map. Or, it is found that an element in the map representing a certain (for convenience of description, called the kth) target information does not exist in the actual road.
  • the i-th vehicle will determine the first prior probability and the second prior probability of the k-th target. Furthermore, sending the first prior probability and the second prior probability to the cloud server may also be referred to as reporting to the cloud server 11 .
  • the cloud server 11 may receive the first prior probability and the second prior probability of the k-th target of the i-th vehicle among the N vehicles through the receiving module.
  • the receiving module can be a data interface or the like.
  • the cloud server 11 may receive the first prior probability and the second prior probability respectively sent by multiple vehicles within a certain time range.
  • the data transmission can be carried out in a self-defined data format, or in a general data format, or in a standard data format.
  • the reporting method can be transmitted by technologies such as network communication and V2X.
  • the network communication technology may include a wireless network, such as wifi, for transmission.
  • the report can be reported according to the first rule, and the report rule can be, for example: whenever a vehicle collects map data corresponding to a certain target or a certain collection area, the report is made, or when the number of data samples collected by a vehicle exceeds When a certain threshold is reached, the report is made, or when a vehicle exceeds a certain threshold according to the number of data samples, the report is made, and so on.
  • the first prior probability is the likelihood conditional probability pi,k(xi
  • the second prior probability is the likelihood conditional probability pi,k(xi
  • the cloud server calculates the first posterior probability and the second posterior probability of the kth target
  • the calculation module of the cloud server 11 calculates the first posteriori of the kth target based on the first and second priori probabilities reported by all vehicles, as well as the preset first and second initial probabilities The probability and the second posterior probability, where the first posterior probability is the posterior probability in the w1 state, the second posterior probability is the posterior probability in the w2 state, the w1 state represents the target existence, and the w2 state represents the target does not. exist.
  • the computing device of the cloud server 11 may specifically be, for example, a processor or the like.
  • the prior probability is the probability that something has not happened yet.
  • the posterior probability is the probability that something has happened, and the reason why it is required to happen is caused by a certain factor.
  • the posterior probability refers to the probability of re-correction after obtaining the information of the "result", which is the "effect” in the problem of "seeking the cause”.
  • the prior probability and the posterior probability are inseparably linked, and the calculation of the posterior probability should be based on the prior probability.
  • the calculation method of the posterior probability can be calculated through a mathematical model. For example, optionally, it can be calculated through the Bayesian formula.
  • the specific formula is as follows:
  • x i k represents the k-th target information observed by the i-th vehicle
  • w 1 ) is the first prior probability for the k-th target information sent by the i-th vehicle
  • P(w1) is the first initial probability
  • the judgment module of the cloud server 11 judges the first posterior probability and the second posterior probability, and if the first posterior probability is greater than the second posterior probability, it is judged that the kth target exists.
  • the first posterior probability is the posterior probability in the w1 state. Since the w1 state represents the existence of the target, the first posterior probability is greater than the second posterior probability, indicating that the kth target has a high probability of existing in the natural environment, that is, the judgment The k-th target exists in the natural environment.
  • the cloud server 11 updates the map data with the k-th target as a newly added element, and the element is an element constituting a map.
  • the update module of the cloud server 11 After determining that the k th target exists in the natural environment, the update module of the cloud server 11 updates the map data, and adds the k th target as a map element to the electronic map as a new element to form a new electronic map.
  • the cloud server 11 can transmit the updated electronic map through wired or wireless network communication, V2X and other technologies, and issue it to the vehicle terminal 12 .
  • the network communication technology may include a wireless network, such as wifi, for transmission.
  • the Bayesian estimation based on the minimum error rate is performed on the vehicle reporting results from different sources, and the posterior probability in the target existence state is compared with the posterior probability in the target nonexistent state, and then a decision is made. It can minimize the error rate of cloud data fusion and improve the accuracy of data fusion.
  • the map data is updated with the k th target as a deletion element.
  • the judgment module of the cloud server 11 judges the first posterior probability and the second posterior probability, and if the first posterior probability is not greater than the second posterior probability, it is judged that the kth target does not exist.
  • the first posterior probability is the posterior probability in the w1 state. Since the w2 state represents that the target does not exist, the first posterior probability is not greater than the second posterior probability, indicating that the kth target has a high probability of not existing in the natural environment. , that is, it is determined that the k-th target does not exist in the natural environment.
  • the cloud server electronic map searches for the target. If the element does not exist in the original electronic map, the cloud server 11 does not update the map data.
  • the cloud server 11 can transmit the updated electronic map through wired or wireless network communication, V2X and other technologies, and issue it to the vehicle terminal 12 .
  • the network communication technology may include a wireless network, such as wifi, for transmission.
  • the first prior probability and the second prior probability are related to the following factors: target type, target location, time, sensor type, design operating condition (ODD) type, lighting conditions, distance, or weather.
  • the probability value is related to the target and the location of the target.
  • the target is a traffic light
  • the prior probability of a traffic light in the intersection area is high.
  • the prior probability is related to time, and the target is still a traffic light. When the intersection is repaired, it will be set to install the traffic light after an average of 3 days. Then the probability of the existence of the traffic light on the first day is lower than the probability of the existence of the traffic light on the third day.
  • a priori probability template may be formed and priori probability quantification may be performed. The specific method will be described in detail below, and will not be repeated here.
  • the embodiment of the present application also provides an apparatus 800 for updating map data.
  • the apparatus specifically includes the following modules:
  • a receiving module 802 configured to receive a first prior probability and a second prior probability sent by the i-th vehicle among the N vehicles for the k-th target, wherein the first prior probability is w
  • the second prior probability is the probability p i,k (x i
  • a calculation module 803 configured to calculate the first a posteriori probability of the kth target based on the first initial probability, the second initial probability, the first prior probability and the second prior probability and the second posterior probability, where the first posterior probability is the posterior probability in the w1 state, and the second posterior probability is the posterior probability in the w2 state;
  • a determination module 804 configured to determine that the kth target exists in the natural environment when the first posterior probability is greater than the second posterior probability.
  • the determining module is further configured to: when the first posterior probability is not greater than the second posterior probability, determine that the kth target does not exist in the natural environment.
  • the updating module 805 is specifically configured to not update the map data when the map does not have elements representing the kth target.
  • the map data is updated by using the k th target as a deletion element.
  • calculation module calculates the first posterior probability and the second posterior probability of the k th target, specifically including the following formulas 1 and 2:
  • x i k represents the k-th target information observed by the i-th vehicle
  • w 1 ) is the first prior probability for the k-th target information sent by the i-th vehicle
  • P(w1) is the first initial probability
  • first initial probability and the second initial probability are preset values.
  • first prior probability and the second prior probability are related to the following factors: target type, target location, time, sensor type, design operating condition (ODD) type, lighting condition, distance, or weather.
  • ODD design operating condition
  • the embodiment of the present application also provides a method for generating a priori probability template, which is used to generate a priori probability template, and the first prior probability and the second prior probability reported by the vehicle terminal 12 can be valued from the template .
  • the prior probability template generation method can be generated by a statistical analysis method, or can also be generated by a theoretical modeling method.
  • the theoretical modeling method first assumes a priori probability distribution model, such as Gaussian distribution, Alpha distribution, and Beta distribution, to obtain a theoretical probability density model, and then corrects the model through experimental values.
  • a priori probability distribution model such as Gaussian distribution, Alpha distribution, and Beta distribution
  • the experimental value may be the target information observed by the vehicle terminal in the process of traveling or when it is parked through sensors, such as lidar, camera millimeter-wave radar, ultrasonic wave, or combined inertial navigation.
  • the target information is: Traffic light information.
  • the vehicle terminal can determine the existence probability of the target information.
  • the existence probability represents the probability of the existence of a certain target observed by the vehicle terminal.
  • the prior probability formation method is related to the following factors: target type, target location, time, sensor type, design operating condition (ODD) type, lighting conditions, distance, or weather.
  • ODD design operating condition
  • the location of the traffic light will affect the size of the prior probability.
  • the new prior probability of the traffic light is high.
  • the probability of existence of traffic lights on the first day is smaller than the probability of existence of traffic lights on the third day. It can be seen that the formation method of the prior probability is related to the target type, target location and time.
  • the specific factors affecting the prior probability are not the same.
  • the influencing factors that affect the prior probability include: different targets, different lighting, different vertical distances (to the target), different horizontal distances (to the target), different weather (haze, sunny, cloudy day).
  • the influencing factors that affect the prior probability include: different targets, reflectivity of different targets, different longitudinal distances (to the target), different lateral distances (to the target), Different weather (haze, sunny, cloudy).
  • the prior probability template may further include a priori probability after fusion of the target information collected by the sensor.
  • the prior probability may also be quantified to form a quantified prior probability template.
  • the quantified prior probability template can be as shown in Table 1 below, where a and b in Table 1 are a real value between (0, 1), a ⁇ b .
  • the quantized prior probability template may be shown in Table 2 below, where a and b in the table below are a real value between (0, 1), a ⁇ b.
  • the quantized prior probability template may be shown in Table 3 below, where a and b in the following table are a real value between (0, 1), a ⁇ b.
  • the quantitative prior probability template may be as shown in Table 4 below, where a and b in the following table are a real value between (0, 1), a ⁇ b.
  • the vehicle terminal sends the first prior probability and the second prior probability to the cloud, and the cloud uses the data information to calculate the first posterior probability and the second posterior probability.
  • the use reliability of the map update data sent to the vehicle terminal is improved, and the safety of the map update data used by the vehicle terminal is guaranteed.
  • the apparatus for updating map data includes:
  • the processor 730 is configured to execute the computer program stored in the memory, so as to implement the data transmission method, the data processing method and the map data update method in the above embodiments. For details, refer to the relevant descriptions in the foregoing method embodiments.
  • the memory 740 may be independent or integrated with the processor 730 .
  • the means for updating map data may further include:
  • a bus for connecting the memory 730 and the processor 740 A bus for connecting the memory 730 and the processor 740 .
  • this embodiment further includes: a communication interface, where the communication interface can be connected to the processor 740 through a bus.
  • the processor 730 can control the communication interface to realize the above-mentioned receiving and sending functions of the map data updating apparatus.
  • the implementation principle and technical effect are similar to those of the method embodiment, and the functions of each module may refer to the corresponding description in the method embodiment, which will not be repeated here.
  • An embodiment of the present application provides a server, as shown in FIG. 5 , which includes the transmission apparatus, the processing apparatus, and the update apparatus of the foregoing embodiment.
  • the implementation principle and technical effect are similar to those of the method embodiment, and the functions of each module may refer to the corresponding description in the method embodiment, which will not be repeated here.
  • An embodiment of the present application further provides a server, and reference may be made to FIG. 5 , and the apparatus may be configured to execute respective steps and/or processes corresponding to the foregoing method embodiments.
  • the implementation principle and technical effect are similar, and the functions of each module may refer to the corresponding description in the method embodiment, which will not be repeated here.
  • An embodiment of the present application further provides a readable storage medium, including: an execution instruction is stored in the readable storage medium, and when at least one processor of a map data update device, a data transmission device, or a data processing device executes the instruction When the instruction is executed, the data processing device, the data transmission device, or the map data update device executes the map data update method described in the above embodiments.
  • a readable storage medium including: an execution instruction is stored in the readable storage medium, and when at least one processor of a map data update device, a data transmission device, or a data processing device executes the instruction
  • the data processing device, the data transmission device, or the map data update device executes the map data update method described in the above embodiments.
  • the implementation principle and technical effect are similar, and the functions of each module may refer to the corresponding description in the method embodiment, which will not be repeated here.
  • Embodiments of the present application further provide a chip, including at least one processor, where the processor is coupled to a memory, and the processor is configured to read instructions in the memory and execute the methods described in the foregoing method embodiments according to the instructions .
  • the implementation principle and technical effect are similar, and the functions of each module may refer to the corresponding description in the method embodiment, which will not be repeated here.
  • An embodiment of the present application further provides a roadside unit, including the updating device of the foregoing embodiment.
  • the implementation principle and technical effect are similar to those of the method embodiment, and the functions of each module may refer to the corresponding description in the method embodiment, which will not be repeated here.
  • example computer program product 600 is provided using signal bearing medium 601 .
  • the signal bearing medium 601 may include one or more program instructions 602 that, when executed by one or more processors, may provide the functions, or portions thereof, described above with respect to FIG. 7 .
  • steps S01 - S08 may be undertaken by one or more instructions associated with the signal bearing medium 601 .
  • program instructions 602 in FIG. 13 also describe example instructions.
  • the signal bearing medium 601 may include a computer-readable medium 603, such as, but not limited to, a hard drive, a compact disc (CD), a digital video disc (DVD), a digital tape, a memory, a read only memory (Read) -Only Memory, ROM) or random access memory (Random Access Memory, RAM) and so on.
  • the signal bearing medium 601 may include a computer recordable medium 604, such as, but not limited to, memory, read/write (R/W) CDs, R/W DVDs, and the like.
  • signal bearing medium 601 may include communication medium 605, such as, but not limited to, digital and/or analog communication media (e.g., fiber optic cables, waveguides, wired communication links, wireless communication links, etc.).
  • the signal bearing medium 601 may be conveyed by a wireless form of communication medium 605 (eg, a wireless communication medium conforming to the IEEE 802.11 standard or other transmission protocol).
  • the one or more program instructions 602 may be, for example, computer-executable instructions or logic-implemented instructions.
  • a computing device such as the computing device described with respect to FIGS.
  • 3-5 may be configured, in response to communication via one or more of computer-readable medium 603 , computer-recordable medium 604 , and/or communication medium 605 , Program instructions 602 communicated to a computing device to provide various operations, functions, or actions.
  • Program instructions 602 communicated to a computing device to provide various operations, functions, or actions.
  • the arrangements described herein are for illustrative purposes only. Thus, those skilled in the art will understand that other arrangements and other elements (eg, machines, interfaces, functions, sequences, and groups of functions, etc.) can be used instead and that some elements may be omitted altogether depending on the desired results . Additionally, many of the described elements are functional entities that may be implemented as discrete or distributed components, or in conjunction with other components in any suitable combination and position.
  • each functional module in the embodiments of the present application may be integrated into one processing module, or each module may exist physically alone, or two or more modules may be integrated into one module.
  • the above-mentioned integrated modules can be implemented in the form of hardware, and can also be implemented in the form of software function modules.
  • the integrated modules are implemented in the form of software functional modules and sold or used as independent products, they may be stored in a computer-readable storage medium.
  • the technical solution of the present application can be embodied in the form of a software product in essence, or a part that contributes to the prior art, or all or part of the technical solution, and the computer software product is stored in a storage inoculation , including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) or a processor to execute all or part of the steps of the methods described in the various embodiments of the present application.
  • the aforementioned storage medium includes: U disk, removable hard disk, ROM, RAM, magnetic disk or optical disk and other media that can store programs.
  • the above-mentioned embodiments it may be implemented in whole or in part by software, hardware, firmware or any combination thereof.
  • software it can be implemented in whole or in part in the form of a computer program product.
  • the computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present application are generated.
  • the computer may be a general purpose computer, special purpose computer, computer network, or other programmable device.
  • the computer instructions may be stored in or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be downloaded from a website site, computer, server, or data center Transmission to another website site, computer, server, or data center is by wire (eg, coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (eg, infrared, wireless, microwave, etc.).
  • the computer-readable storage medium can be any available ring that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains one or more of the available media integrations.
  • the available media may be magnetic media (eg, floppy disks, hard disks, magnetic tapes), optical media (eg, DVDs), or semiconductor media (eg, solid State Disk (SSD), etc.).
  • the program instructions can be implemented in the form of software functional units and can be sold or used as a stand-alone product, and the memory can be any form of computer-readable storage medium.
  • the memory can be any form of computer-readable storage medium.
  • all or part of the technical solutions of the present application may be embodied in the form of software products, including several instructions to enable hundreds of millions of computer devices, specifically processors, to execute the target detection device in each embodiment of the present application. all or part of the steps.
  • the aforementioned computer-readable storage medium includes: U disk, removable hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disks or optical disks and other programs that can store programs medium.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Theoretical Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Mathematical Physics (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Analysis (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Optimization (AREA)
  • Computational Mathematics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Operations Research (AREA)
  • Probability & Statistics with Applications (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Algebra (AREA)
  • Remote Sensing (AREA)
  • Software Systems (AREA)
  • Traffic Control Systems (AREA)

Abstract

本申请提供一种数据传输方法，所述方法包括：确定目标信息对应的地图元素的至少一个先验概率，其中，所述目标信息是车辆感知周边环境所检测到的关于目标的信息，所述先验概率是所述目标在第一条件下存在的概率，所述至少一个先验概率中的每个先验概率对应一个所述第一条件；发送所述目标信息、所述至少一个先验概率以及与所述每个先验概率对应的第一条件。

Description

一种地图数据的传输方法及装置
相关申请的交叉引用
本申请要求在2020年09月29日提交中国专利局、申请号为202011054715.7、申请名称为“一种地图数据的传输方法及装置”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及数据传输技术领域，尤其涉及一种地图数据的传输方法及装置。
背景技术
自动驾驶汽车依靠人工智能、视觉计算、雷达、定位系统和高精度地图等技术的协同合作，让电脑可以在没有任何人类主动的操作下，自动安全地操作机动车辆。高精度地图作为汽车导航使用的工具，其准确度和精度以及更新的效率对于自动驾驶汽车的安全至关重要，比如当道路被封闭、路线发生变化或者交通标示有改变时，需要更新地图。
目前，众包作为一种成本较低的数据采集模式近几年来被广泛采用，众包就是基于大众的力量完成某种特定工作任务，例如，基于大量的大众车辆完成地图数据的采集。车辆可以上报每个传感器的检测信息到车内融合单元，但上报信息是传感器特定的形式，即，不同传感器有不同的上报内容和格式。或者，车辆上报传感器检测到的目标结果信息到云端。注意，这里检测结果信息只支持目标这一种类型，其中，目标是自然环境中的物体。众包地图更新中，由各个车辆上报检测到的目标信息，在上报过程中，一般也会上报置信度信息，用于判断目标信息的可信程度。
云端拿到该置信度值后可以对多辆车的上报结果进行融合，目前业界最常用的融合方法包括基于最小错误率的贝叶斯决策等方法，在该方法中，云端可以基于多辆车的检测结果计算得到待检测目标的后验概率并将该概率值作为最终的置信度值。
但是上述技术中的贝叶斯决策的缺陷包括：如图1所示，先验假设正确度的依赖性过高，当先验假设正确时，置信度曲线没有超出可信的上限，置信度曲线是可信的。但是当先验假设不正确时，置信度曲线超出了可信的上限，置信度曲线不能正确反映置信度与错误率的关系，即置信度的估计可能是无效的，此时就会导致云端对目标是否存在的判断产生错误。
发明内容
本申请实施例提供一种地图数据的更新方法及装置，可最大程度减少云端数据融合的错误率，提高数据融合的精度。
为达到上述目的，本申请的实施例采用如下技术方案：
第一方面，提供了一种数据传输方法，所述方法包括：确定目标信息对应的地图元素的至少一个先验概率，其中，所述目标信息是车辆感知周边环境所检测到的关于 目标的信息，所述先验概率是所述目标在第一条件下存在的概率，所述至少一个先验概率中的每个先验概率对应一个所述第一条件；
发送所述目标信息、所述至少一个先验概率以及与所述每个先验概率对应的第一条件。
在第一方面的一种可能的实现方式中：所述第一条件为所述目标信息对应的地图元素的类型、地图元素的位置、地图元素的检测时间、传感器类型、或设计运行条件中的至少一个。
在第一方面的一种可能的实现方式中：所述传感器类型包括激光雷达、摄像头、毫米波雷达、或超声波中的至少一个。
在第一方面的一种可能的实现方式中：所述对应关系包括表格或公式。
在第一方面的一种可能的实现方式中：不同的所述传感器类型对应的所述第一条件不同。
在第一方面的一种可能的实现方式中：当所述传感器类型为摄像头时，所述第一条件包括所述目标信息对应的光照条件、所述目标信息对应的纵向距离、所述目标信息对应的横向距离、或所述目标信息对应的天气信息中的至少一个。
在第一方面的一种可能的实现方式中：当所述传感器类型为激光雷达时，所述先验概率对应的第一条件包括所述目标信息对应的反射率、所述目标信息对应的纵向距离、所述目标信息对应的横向距离、或所述目标信息对应的天气信息中的至少一个。
在第一方面的一种可能的实现方式中：当所述传感器类型为毫米波雷达时，所述先验概率对应的第一条件包括所述目标信息对应的干扰级别、所述目标信息对应的纵向距离、所述目标信息对应的横向距离中的至少一个。
在第一方面的一种可能的实现方式中：所述先验概率为贝叶斯估计算法中的先验概率的量化值。
在上述技术方案中，综合考虑了多种影响因素，对于不同的目标类型考虑不同的影响目标是否存在的因素，从而提高了上报的目标信息先验概率的可靠性。
第二方面，提供了一种数据传输装置，所述装置包括：
确定模块，用于确定目标信息对应的地图元素的至少一个先验概率，其中，所述目标信息是车辆感知周边环境所检测到的关于目标的信息，所述先验概率是所述目标在第一条件下存在的概率，所述至少一个先验概率中的每个先验概率对应一个所述第一条件；
发送模块，用于发送所述目标信息、所述至少一个先验概率以及与所述每个先验概率对应的第一条件。
在第二方面的一种可能的实现方式中：所述第一条件包括所述目标信息对应的地图元素的类型、地图元素的位置、地图元素的检测时间、传感器类型、或设计运行条件中的至少一个。
在第二方面的一种可能的实现方式中：所述传感器类型包括激光雷达、摄像头、毫米波雷达、或超声波中的至少一个。
在第二方面的一种可能的实现方式中：所述对应关系包括表格或公式。
在第二方面的一种可能的实现方式中：不同的所述传感器类型对应的所述第一条 件不同。
在第二方面的一种可能的实现方式中：当所述传感器类型为摄像头时，所述第一条件包括所述目标信息对应的光照条件、所述目标信息对应的纵向距离、所述目标信息对应的横向距离、或所述目标信息对应的天气信息中的至少一个。
在第二方面的一种可能的实现方式中：当所述传感器类型为激光雷达时，所述先验概率对应的第一条件包括所述目标信息对应的反射率、所述目标信息对应的纵向距离、所述目标信息对应的横向距离、或所述目标信息对应的天气信息中的至少一个。
在第二方面的一种可能的实现方式中：当所述传感器类型为毫米波雷达时，所述先验概率对应的第一条件包括所述目标信息对应的干扰级别、所述目标信息对应的纵向距离、所述目标信息对应的横向距离中的至少一个。
在第二方面的一种可能的实现方式中：所述先验概率为贝叶斯估计算法中的先验概率的量化值。
第三方面，本申请提供了一种数据处理方法，所述方法包括：
接收目标信息、至少一个先验概率以及与所述至少一个先验概率中的每个先验概率对应的第一条件，其中，所述目标信息为车辆感知周边环境所检测到的关于目标的信息，所述先验概率是所述目标在第一条件下存在的概率，所述至少一个先验概率中的每个先验概率对应一个所述第一条件；
基于所述至少一个先验概率计算所述目标信息的后验概率；
基于所述后验概率,判定所述目标在所述自然环境中是否存在。
在第三方面的一种可能的实现方式中：所述接收至少一个先验概率包括：
接收第一先验概率和第二先验概率，其中，所述第一先验概率是在w1状态下的先验概率，所述w1表征所述目标存在，所述第二先验概率是w2状态下的先验概率，所述w2表征所述目标不存在。
在第三方面的一种可能的实现方式中：所述基于所述至少一个先验概率计算所述目标信息的后验概率包括：
基于所述第一先验概率计算第一后验概率；
基于所述第二先验概率计算第二后验概率；
其中第一后验概率是所述w1状态下的后验概率，第二后验概率是所述w2状态下的后验概率。
在第三方面的一种可能的实现方式中：所述基于所述后验概率,判定所述目标在所述自然环境中是否存在包括：
当所述第一后验概率大于所述第二后验概率时，判定所述目标在所述自然环境中存在；或者
当所述第一后验概率不大于所述第二后验概率时，判定所述目标在所述自然环境中不存在。
在第三方面的一种可能的实现方式中：所述先验概率为贝叶斯估计算法中的先验概率的量化值；
或者
所述后验概率为贝叶斯估计算法中的后验概率的量化值。
在上述技术方案中，对车辆上报结果执行基于最小错误率的贝叶斯估计，对目标存在状态下的后验概率和目标不存在状态下的后验概率进行比较，进而判定决策。可最大程度减少云端数据融合的错误率，提高数据融合的精度。
第四方面，本申请提供了一种数据处理装置，所述装置包括：
接收模块，用于接收目标信息、至少一个先验概率以及与所述至少一个先验概率中的每个先验概率对应的第一条件，其中，所述目标信息反映了车辆感知周边环境所检测到的关于目标的信息，所述先验概率是所述目标在第一条件下存在的概率，所述至少一个先验概率中的每个先验概率对应一个所述第一条件；
计算模块，用于基于所述至少一个先验概率计算所述目标信息的后验概率；
判定模块，用于基于所述后验概率,判定所述目标在所述自然环境中是否存在。
在第三方面的一种可能的实现方式中：所述接收模块用于：
接收第一先验概率和第二先验概率，其中，所述第一先验概率是在w1状态下的先验概率，所述w1表征所述目标存在，所述第二先验概率是w2状态下的先验概率，所述w2表征所述目标不存在。
在第三方面的一种可能的实现方式中：所述计算模块具体用于：
基于所述第一先验概率计算第一后验概率；
基于所述第二先验概率计算第二后验概率；
其中第一后验概率是所述w1状态下的后验概率，第二后验概率是所述w2状态下的后验概率。
在第三方面的一种可能的实现方式中：所述判定模块具体用于：
当所述第一后验概率大于所述第二后验概率时，判定所述目标在所述自然环境中存在；或者
当所述第一后验概率不大于所述第二后验概率时，判定所述目标在所述自然环境中不存在。
在第三方面的一种可能的实现方式中：所述先验概率为贝叶斯估计算法中的先验概率的量化值；
或者
所述后验概率为贝叶斯估计算法中的后验概率的量化值。
第五方面，提供了一种数据传输装置，包括：存储器，用于存储程序指令；
处理器，用于当调用并执行存储器中的程序指令时，执行第一方面或第一方面的任一种可能的所述的数据传输方法。
第六方面，提供了一种可读存储介质，包括：所述可读存储介质中存储有执行指令，当数据传输装置的至少一个处理器执行该执行指令时，所述数据的传输装置执行第一方面或第一方面的任一种可能的所述的数据传输的方法。
第七方面，提供了一种数据处理装置，包括：存储器，用于存储程序指令；
处理器，用于当调用并执行存储器中的程序指令时，执行第三方面或第三方面的任一种可能的所述的数据处理方法。
第八方面，提供了一种可读存储介质，包括：所述可读存储介质中存储有执行指令，当数据传输装置的至少一个处理器执行该执行指令时，所述数据的传输装置执行 第三方面或第三方面的任一种可能的所述的数据处理的方法。
第九方面，提供了一种服务器，包括第四方面或第四方面的任一种可能的所述的数据处理装置。
第十方面，提供了一种芯片，包括至少一个处理器，所述处理器与存储器耦合，所述处理器用于读取存储器中的指令并根据所述指令执行第一方面或第三方面任一项所述的方法。
第十一方面，提供了一种计算机程序产品，当其在计算机上运行时，使得计算机执行上述第一方面或第四方面任一项所述的方法。
第十二方面，提供了一种路侧单元，包括第二方面或第四方面任一种可能的所述装置。
可以理解地，上述提供的任一种数据处理装置、可读存储介质、计算机程序产品、服务器、芯片、路侧单元，均可以由上文所提供的对应的方法来实现，因此，其所能达到的有益效果可参考上文所提供的对应的方法中的有益效果，此处不再赘述。
附图说明
图1为现有技术中根据置信度进行融合的贝叶斯决策曲线；
图2为本申请实施例提供的一种地图数据处理方法应用场景示意图；
图3为本申请实施例提供的另一种地图数据处理方法应用场景示意图；
图4为本申请实施例提供的再一种地图数据处理方法应用场景示意图；
图5为本申请实施例提供的指令侧地图数据更新结构示意图；
图6为本申请实施例提供的车端观测的目标的示意图；
图7为本申请实施例提供的一种数据传输方法流程图；
图8为本申请实施例提供的一种数据传输模块结构示意图；
图9为本申请实施例提供的一种数据处理方法流程图；
图10为本申请实施例提供的一种数据处理装置结构示意图；
图11为本申请实施例提供的一种地图数据的更新方法；
图12为本申请实施例提供的一种地图数据的更新装置；
图13为本申请实施例提供的计算机程序产品的结构示意图。
具体实施方式
以下，术语“第一”、“第二”仅用于描述目的，而不能理解为指示或暗示相对重要性或者隐含指明所指示的技术特征的数量。由此，限定有“第一”、“第二”的特征可以明示或者隐含地包括一个或者更多个该特征。在本实施例的描述中，除非另有说明，“多个”的含义是两个或两个以上。
同时，在本申请实施例中，“示例性的”或者“例如”等词用于表示作例子、例证或说明。本申请实施例中被描述为“示例性的”或者“例如”的任何实施例或设计方案不应被解释为比其它实施例或设计方案更优选或更具优势。确切而言，使用“示例性的”或者“例如”等词旨在以具体方式呈现相关概念，便于理解。
本申请提供了一种数据传输方法，在车辆终端确定目标信息对应的地图元素的至 少一个先验概率，其中，所述目标信息是车辆感知周边环境所检测到的关于目标的信息，所述先验概率是所述目标在第一条件下存在的概率，所述至少一个先验概率中的每个先验概率对应一个所述第一条件；所述车辆终端向云端服务器发送所述目标信息、所述至少一个先验概率以及与所述每个先验概率对应的第一条件。在上述技术方案中，综合考虑了多种影响因素，对于不同的目标类型考虑不同的影响目标是否存在的因素，从而提高了车辆终端上报的目标信息先验概率的可靠性。
本申请还提供一种地图数据的处理方法和装置，云端服务器从车辆终端接收目标信息、至少一个先验概率以及与所述至少一个先验概率中的每个先验概率对应的第一条件，其中，所述目标信息为车辆感知周边环境所检测到的关于目标的信息，所述先验概率是所述目标在第一条件下存在的概率，所述至少一个先验概率中的每个先验概率对应一个所述第一条件；基于所述至少一个先验概率计算所述目标信息的后验概率；基于所述后验概率,判定所述目标在所述自然环境中是否存在，如果存在将对高精度地图进行更新。由于云端服务器11对不同来源的车辆上报结果执行基于最小错误率的贝叶斯估计，对目标存在状态下的后验概率和目标不存在状态下的后验概率进行比较，进而判定决策，可最大程度减少云端数据融合的错误率，提高数据融合的精度。
本申请提供的地图数据的处理方法、装置及系统，可应用于无人驾驶(unmanned driving)、辅助驾驶(driver assistance/ADAS)或智能驾驶(intelligent driving)领域中。
本申请中的地图可以是电子地图，电子地图即数字地图，包括高精度地图。电子地图是以地图数据库为基础，利用计算机技术，以数字形式存储，可以在终端设备的屏幕上显示的地图。电子地图的主要构成元素就是地图元素，例如山脉、水系、陆地、行政区划、兴趣点或者道路等地理元素，再例如车道线、人行横道、停止线、交通标志、路标、灯杆、红绿灯、龙门架、环岛、停车场等道路上的目标元素，其中，道路还可以进一步划分为高速公路、一级公路、二级公路、三级公路和四级公路五个等级，每个等级的道路可以为不同的地图元素。
云端服务器11可以通过无线网络为多个车辆12提供实时的地图数据，该云端服务器11包括较大容量的存储空间，用于存储地图数据，包括高精度地图，并且负责将电子地图更新下发等。具体的，可以将地图数据部署于一台或者多台服务器上。可选的，在云端服务器12中可以由网络平台基于众包车辆终端12上报的道路数据或者更新的目标地图元素，决策是否对当前地图进行更新，并执行对地图数据的更新工作，更新后可以下发新的电子地图。
车辆终端12是用于车辆通信和管理的前端设备，可以安装在各种车辆内。至少一个车辆终端12可以是大规模的社会车辆中的通信装置，也可以是具有数据处理能力和识别能力的智能车辆中的通信装置，还可以由普通社会车辆和智能车辆组成。众包的车辆终端12可以作为地图采集和更新的重要基础，可以通过车辆终端中配置的车载传感器获取实时的道路检测信息，例如，通过摄像头，红外传感器、雷达探测仪、全球卫星导航系统或者惯性导航系统(简称惯导)等，可以将检测的道路信息上传到云端服务器11进行地图数据的实时更新和维护。具体的，车辆终端12可以通过网络将采集到的道路数据或者更新的目标地图元素上传至云端服务器11或者其他网 络设备。
在一种实施例中，该通信系统还可以包括路侧单元13，通过有线或者无线网络与车辆终端11或者云端服务器12通信，可以向车辆终端11提供一定区域范围内的道路信息、高精度的定位、或者提供高精度地图等服务。路侧单元13还可以用于收集一定区域范围内的车辆终端11上报的道路检测信息，经过数据处理和融合，确定相应的地图元素是否发生变化，从而决策是否更新地图。
图2为本申请提供的地图数据的处理方法应用的一种场景示意图，如图2所示，在该应用场景下，主要涉及云端服务器11和车辆终端12。其中，车辆终端12使用本地地图，并可以感知周边环境，例如可以通过设置在车辆终端12上的传感器感知周边环境，检测到目标信息，车辆终端12根据检测到的目标信息确定出先验概率。云端服务器11上可以存储有高精度地图数据库，云端服务器11可以接收至少两个来自车辆终端12的目标信息，云端设置接收目标信息、至少一个先验概率以及与所述至少一个先验概率中的每个先验概率对应的第一条件，基于所述至少一个先验概率计算所述目标信息的后验概率，基于所述后验概率,判定所述目标在所述自然环境中是否存在，进一步判断是否更新地图。
图3为本申请提供的地图数据的处理方法应用的另一种场景示意图，如图3所示，在该应用场景下，主要涉及云端服务器11、车辆终端12和基站或路边单元设备13，其中，车辆终端12使用本地地图，并可以感知周边环境，例如可以通过设置在车辆终端12上的传感器感知周边环境，检测到目标信息，车辆终端12根据检测到的目标信息确定出先验概率。云端服务器11上可以存储有高精度地图数据库，云端服务器11可以接收至少两个来自车辆终端12的目标信息，云端服务器11可以接收至少两个来自车辆终端12的目标信息，云端设置接收目标信息、至少一个先验概率以及与所述至少一个先验概率中的每个先验概率对应的第一条件，基于所述至少一个先验概率计算所述目标信息的后验概率，基于所述后验概率,判定所述目标在所述自然环境中是否存在，进一步判断是否更新地图。。云端服务器11将高精地图进行更新并下发，路边单元设备(road side unit，RSU)13根据云端服务器11下发的指令对不匹配的目标信息进行重点感知，并将感知结果进行上报。
图4为本申请提供的一种地图数据更新方法应用的再一种场景示意图，如图4所示，在该场景下，主要涉及车辆终端12和基站或路边单元设备13，其中，车辆终端12使用本地地图，并可以感知周边环境，例如可以通过设置在车辆终端12上的传感器感知周边环境，检测到目标信息，车辆终端12根据检测到的目标信息确定出先验概率。车辆终端12将所述先验概率发送给基站或路边单元设备13，路边单元设备13接收目标信息、至少一个先验概率以及与所述至少一个先验概率中的每个先验概率对应的第一条件，基于所述至少一个先验概率计算所述目标信息的后验概率，基于所述后验概率,判定所述目标在所述自然环境中是否存在，进一步判断是否更新地图。图5为本申请实施例提供的一种云端指令侧地图数据更新结构示意图。
车端计算机系统112还可以从其它计算机系统接收信息或转移信息到其它计算机系统。或者，从车辆终端12的传感器系统收集的传感器数据可以被转移到另一个计算机对此数据进行处理。如图5所示，来自计算机系统112的数据可以经由网络被传送 到云侧的计算机720用于进一步的处理。网络以及中间节点可以包括各种配置和协议，包括因特网、万维网、内联网、虚拟专用网络、广域网、局域网、使用一个或多个公司的专有通信协议的专用网络、以太网、WiFi和HTTP、以及前述的各种组合。这种通信可以由能够传送数据到其它计算机和从其它计算机传送数据的任何设备，诸如调制解调器和无线接口。
在一个示例中，计算机720可以包括具有多个计算机的服务器，例如负载均衡服务器群，为了从计算机系统112接收、处理并传送数据的目的，其与网络的不同节点交换信息。该服务器可以被类似于计算机系统110配置，具有处理器730、存储器740、指令750、和数据760。
数据760可以包括观测目标的先验概率，服务器720可以接受、监视、存储、更新、以及与地图数据相关的各种信息，并且判别地图数据是否更新。
下面结合附图详细说明本申请的技术方案。图7为本申请实施例提供的一种数据传输方法，如图7所示，本实施例的执行主体为车辆。
在本申请中，具体包括以下步骤：
S101、确定目标信息对应的地图元素的至少一个先验概率，其中，所述目标信息是车辆感知周边环境所检测到的关于目标的信息，所述先验概率是所述目标在第一条件下存在的概率，所述至少一个先验概率中的每个先验概率对应一个所述第一条件。
其中，如图6所示，目标可以是自然环境中的物体，具体为实际道路中影响车辆终端行驶的物体，例如为道路旁边的红绿灯，道路两侧的建筑物，或者为路标、灯杆、交通标志牌等。地图元素为组成地图的要素，电子地图包括多个地图元素。
目标信息可以为车辆终端感知周边环境所检测到的关于目标信息，例如车辆终端在运行的过程中，通过设置在车辆终端上的传感器，如，激光雷达、摄像头毫米波雷达、超声波或组合惯导等传感器检测周围的环境得到目标信息。车辆终端12将检测到的目标信息发送给云端服务器11。
先验概率为在第一条件下目标存在的概率，第一条件为所述目标信息对应的地图元素的种类、地图元素的位置、地图元素的检测时间、所述目标信息对应的传感器类型、或所述目标信息对应的设计运行条件(ODD)类型中的至少一个。
车端可以上报多个先验概率，所述多个先验概率分别具有多个不同的第一条件，例如道路旁边的红绿灯在路口存在的先验概率；在路口维修的情况下，假设平均三天安装红绿灯，在三天后存在的先验概率等。其中所述至少一个先验概率中的每个先验概率对应一个所述第一条件。
S102、发送所述目标信息、所述至少一个先验概率以及与所述每个先验概率对应的第一条件。
其中，车辆终端需要向云端发送三项数据，包括该目标信息、需要上报的先验概率、以及该先验概率对应的第一条件。先验概率可以为一个或多个，相应的第一条件也对应的为一个或多个。
数据的传输可以采用自定义的数据格式，或者采用通用的数据格式，或者采用标准的数据格式来进行传输。上报方法可以网络通信、V2X等技术进行传输。网络通信技术可以包括无线网络，例如wifi等方式进行传输方式。
可选的，该第一条件为所述目标信息对应的地图元素的类型、地图元素的位置、地图元素的检测时间、传感器类型、或设计运行条件中的至少一个。
可选的，该传感器类型包括激光雷达、摄像头、毫米波雷达、或超声波中的至少一个。
可选的，该对应关系包括表格或公式，其中表格可以参见下文的表格1-4所示。
可选的，不同的所述传感器类型对应的所述第一条件不同。
可选的，当所述传感器类型为摄像头时，所述第一条件包括所述目标信息对应的光照条件、所述目标信息对应的纵向距离、所述目标信息对应的横向距离、或所述目标信息对应的天气信息中的至少一个。
可选的，当所述传感器类型为激光雷达时，所述先验概率对应的第一条件包括所述目标信息对应的反射率、所述目标信息对应的纵向距离、所述目标信息对应的横向距离、或所述目标信息对应的天气信息中的至少一个。
可选的，当所述传感器类型为毫米波雷达时，所述先验概率对应的第一条件包括所述目标信息对应的干扰级别、所述目标信息对应的纵向距离、所述目标信息对应的横向距离中的至少一个。
可选的，所述先验概率为贝叶斯估计算法中的先验概率的量化值。
在上述技术方案中，使用不同条件的先验概率，综合考虑了多种影响因素，对于不同的目标类型考虑不同的影响目标是否存在的因素，从而提高了车辆终端上报的目标信息先验概率的可靠性。
本申请还提供了一种数据发送装置200，如图8所示，主要包括以下模块：
确定模块201，用于确定目标信息对应的地图元素的至少一个先验概率，其中，所述目标信息是车辆感知周边环境所检测到的关于目标的信息，所述先验概率是所述目标在第一条件下存在的概率，所述至少一个先验概率中的每个先验概率对应一个所述第一条件；
发送模块202，用于发送所述目标信息、所述至少一个先验概率以及与所述每个先验概率对应的第一条件。
可选的，所述第一条件为所述目标信息对应的地图元素的类型、地图元素的位置、地图元素的检测时间、传感器类型、或设计运行条件中的至少一个。
可选的，所述传感器类型包括激光雷达、摄像头、毫米波雷达、或超声波中的至少一个。
可选的，所述对应关系包括表格或公式，其中表格可以参见下文的表格1-4所示。
可选的，不同的所述传感器类型对应的所述第一条件不同。
可选的，当所述传感器类型为摄像头时，所述第一条件包括所述目标信息对应的光照条件、所述目标信息对应的纵向距离、所述目标信息对应的横向距离、或所述目标信息对应的天气信息中的至少一个。
可选的，当所述传感器类型为激光雷达时，所述先验概率对应的第一条件包括所述目标信息对应的反射率、所述目标信息对应的纵向距离、所述目标信息对应的横向距离、或所述目标信息对应的天气信息中的至少一个。
可选的，当所述传感器类型为毫米波雷达时，所述先验概率对应的第一条件包括 所述目标信息对应的干扰级别、所述目标信息对应的纵向距离、所述目标信息对应的横向距离中的至少一个。
可选的，所述先验概率为贝叶斯估计算法中的先验概率的量化值。
需要说明的是，上述数据传输装置与执行上述传输方法实施例的技术方案，其实现原理和技术效果类似，此处不再赘述。
图9为本申请提供的一种数据的处理方法流程图，如图9所示，本实施例的执行主体可以为云端服务器11或者基站或路边单元设备13，下面以云端服务器11为例进行描述。该数据的处理方法具体包括以下几个步骤：
S301、接收目标信息、至少一个先验概率以及与所述至少一个先验概率中的每个先验概率对应的第一条件，其中，所述目标信息为车辆感知周边环境所检测到的关于目标的信息，所述先验概率是所述目标在第一条件下存在的概率，所述至少一个先验概率中的每个先验概率对应一个所述第一条件。
其中，云端服务器11可以为地图云，具体的动作执行对象可以为云端服务器中的计算装置，如处理器等。
其中云端服务器11接收的数据为车辆终端按照一定的数据格式发送的数据。该数据包括三项内容：目标信息、该目标信息的先验概率、以及该先验概率对应的第一条件。该先验概率可以为基于不同的第一条件的多个先验概率。
云端服务器接收数据的格式可以采用自定义的数据格式，或者采用通用的数据格式，或者采用标准的数据格式来进行传输。上报方法可以网络通信、V2X等技术进行传输。网络通信技术可以包括无线网络，例如wifi等方式进行传输方式。
S302、基于所述至少一个先验概率计算所述目标信息的后验概率。
其中，先验概率是事情还没有发生，要求这件事情发生的可能性的大小。后验概率是事情已经发生，要求这件事情发生的原因是由某个因素引起的可能性的大小。后验概率是指在得到“结果”的信息后重新修正的概率，是“执果寻因”问题中的"果"。先验概率与后验概率有不可分割的联系，后验概率的计算要以先验概率为基础。
可选的，后验概率的计算方法可以通过数学模型来进行计算，例如，可选的，可以通过贝叶斯公式进行计算。可选的，所述先验概率为贝叶斯估计算法中的先验概率的量化值；或者所述后验概率为贝叶斯估计算法中的后验概率的量化值。
可选的，可以基于所述第一先验概率计算第一后验概率；基于所述第二先验概率计算第二后验概率；其中第一后验概率是所述w1状态下的后验概率，第二后验概率是所述w2状态下的后验概率。
S303、基于所述后验概率,判定所述目标在所述自然环境中是否存在。
其中，云端服务器11的判定模块基于后验概率的结果来判定该目标在自然环境中是否存在。
可选的，当所述第一后验概率大于所述第二后验概率时，判定所述目标在所述自然环境中存在；或者
当所述第一后验概率不大于所述第二后验概率时，判定所述目标在所述自然环境中不存在。
在上述技术方案中，对不同来源的车辆上报结果执行基于最小错误率的贝叶斯估 计，对目标存在状态下的后验概率和目标不存在状态下的后验概率进行比较，进而判定决策。可最大程度减少云端数据融合的错误率，提高数据融合的精度。
图10为本申请提供的一种数据处理装置400的结构示意图，所述数据处理装置400主要包括以下几个模块：
接收模块401，用于接收目标信息、至少一个先验概率以及与所述至少一个先验概率中的每个先验概率对应的第一条件，其中，所述目标信息反映了车辆感知周边环境所检测到的关于目标的信息，所述先验概率是所述目标在第一条件下存在的概率，所述至少一个先验概率中的每个先验概率对应一个所述第一条件；
计算模块402，用于基于所述至少一个先验概率计算所述目标信息的后验概率；
判定模块403，用于基于所述后验概率,判定所述目标在所述自然环境中是否存在。
进一步的，所述接收模块用于：接收第一先验概率和第二先验概率，其中，所述第一先验概率是在w1状态下的先验概率，所述w1表征所述目标存在，所述第二先验概率是w2状态下的先验概率，所述w2表征所述目标不存在。
进一步的，所述计算模块具体用于：基于所述第一先验概率计算第一后验概率；基于所述第二先验概率计算第二后验概率；其中第一后验概率是所述w1状态下的后验概率，第二后验概率是所述w2状态下的后验概率。
进一步的，所述判定模块具体用于：当所述第一后验概率大于所述第二后验概率时，判定所述目标在所述自然环境中存在；或者当所述第一后验概率不大于所述第二后验概率时，判定所述目标在所述自然环境中不存在。
进一步的，所述先验概率为贝叶斯估计算法中的先验概率的量化值；或者所述后验概率为贝叶斯估计算法中的后验概率的量化值。
需要说明的是，上述数据传输装置与执行上述传输方法实施例的技术方案，其实现原理和技术效果类似，此处不再赘述。
如图11本申请还提供了一种地图数据更新方法。在本申请中，假设各车的观测是独立的。下面介绍本实施例的地图数据的更新方法，具体包括以下步骤：
S01：云端服务器11确定第k个目标的第一初始概率和第二初始概率。
其中，云端服务器11可以为地图云，具体的动作执行对象可以为云端服务器中的计算装置，如处理器等。
如图6所示，目标可以是自然环境中的物体，具体为实际道路中影响车辆终端行驶的物体，例如为道路旁边的红绿灯，道路两侧的建筑物，或者为路标、灯杆、交通标志牌等。地图元素为组成地图的要素，电子地图包括多个地图元素。
目标信息可以为车辆终端感知周边环境所检测到的关于目标信息，例如车辆终端在运行的过程中，通过设置在车辆终端上的传感器，如，激光雷达、摄像头毫米波雷达、超声波或组合惯导等传感器检测周围的环境得到目标信息。车辆终端12将检测到的目标信息上报给云端服务器11，该服务器11判断目标存在与否，若果存在，将该目标作为新增的地图元素添加到地图中。
可见，电子地图中包括多个元素，每个元素代表一个目标，第k个目标为其中的一个目标，k为正整数。
云端服务器可以预先设置第一初始概率P(w 1)＝a，w 1表征目标存在，第二初始概 率P(w 2)＝1-a，w 2表征目标不存在。
该第一初始概率P(w 1)＝a和该第二初始概率P(w 2)＝1-a可以根据是本领域的技术人员根据经验来设置，也可以是实验人员根据多次试验的结果进行设置，该值是可以动态优化和调整的。
S02：云端服务器11接收所述N个车辆中的第i个车辆针对第k个目标的第一先验概率和第二先验概率。
其中，参与众包的车辆终端12可以包含N个车辆，N为大于等于2的正整数，也就是说至少两个车辆参与目标信息的采集。N个车辆中的第i个车辆在行进的过程中或者停靠的时候，通过传感器，如激光雷达、摄像头毫米波雷达、超声波或组合惯导等传感器观测到第k个目标，发现该目标的信息与地图中的信息并不一致，比如，在地图中未有表示该目标的元素信息。或者，发现地图中的表征某个(为了描述方便，称为第k个)目标信息的元素，在实际道路中并不存在。总是，当实际道路与电子地图的信息不一致时，第i个车辆将确定该第k个目标的第一先验概率和第二先验概率。并且，将该第一先验概率和该第二先验概率发送给云端服务器，也可以称为上报给云端服务器11。
云端服务器11可以通过接收模块接收所述N个车辆中的第i车辆针对第k个目标的第一先验概率和第二先验概率。接收模块可以为数据接口等。
云端服务器11可以在一定的时间范围内接收多个车辆分别发送的第一先验概率和第二先验概率。
数据的传输可以采用自定义的数据格式，或者采用通用的数据格式，或者采用标准的数据格式来进行传输。上报方法可以网络通信、V2X等技术进行传输。网络通信技术可以包括无线网络，例如wifi等方式进行传输方式。
上报可以按照第一规则进行上报，上报规则例如可以是：每当一个车辆采集到某一目标或某一采集区域对应的地图数据时即进行上报，或者，当一个车辆采集到的数据样本数超过某一阈值时即进行上报，或者，当一个车辆根据数据样本数超过某一阈值时即进行上报等等。
该第一先验概率为w1状态下的似然条件概率pi,k(xi|w1)，所述第二先验概率为w2状态下的似然条件概率pi,k(xi|w2)，1≤i≤N，N≥2且为整数。
S03：云端服务器计算第k个目标的第一后验概率和第二后验概率
其中，云端服务器11的计算模块基于所有车辆分别上报的第一先验概率和第二先验概率，以及预设的第一初始概率、第二初始概率计算该第k个目标的第一后验概率和第二后验概率，其中第一后验概率是w 1状态下的后验概率，第二后验概率是w2状态下的后验概率，w 1状态表征目标存在，w2状态表征目标不存在。云端服务器11的计算装置具体可以为例如处理器等。
先验概率是事情还没有发生，要求这件事情发生的可能性的大小。后验概率是事情已经发生，要求这件事情发生的原因是由某个因素引起的可能性的大小。后验概率是指在得到“结果”的信息后重新修正的概率，是“执果寻因”问题中的"果"。先验概率与后验概率有不可分割的联系，后验概率的计算要以先验概率为基础。
后验概率的计算方法可以通过数学模型来进行计算，例如，可选的，可以通过贝 叶斯公式进行计算，具体公式如下：
Figure PCTCN2021117860-appb-000001
Figure PCTCN2021117860-appb-000002
其中，
Figure PCTCN2021117860-appb-000003
为第k个目标信息的第一后验概率，x i k表示第i个车辆观测到第k个目标信息，
p i，k(x i|w 1)为第i个车辆发送的针对第k个目标信息的第一先验概率，P(w1)为第一初始概率,
Figure PCTCN2021117860-appb-000004
为第k个目标信息的第二后验概率，p i，k(x i|w 2)为第i个车辆发送的针对第k个目标信息的第一先验概率，P(w2)为第二初始概率。
S04：当所述第一后验概率大于所述第二后验概率时,判定所述第k个目标在所述自然环境中存在。
云端服务器11的判断模块对该第一后验概率和第二后验概率进行判定，如果第一后验概率大于第二后验概率,判定第k个目标存在。第一后验概率是w1状态下的后验概率，由于w 1状态表征目标存在，第一后验概率大于第二后验概率说明该第k个目标在自然环境中存在的概率大，即判定第k个目标在所述自然环境中存在。
S05：云端服务器11将第k个目标作为新增元素更新地图数据，所述元素为构成地图的要素。
当判定第k个目标在自然环境中存在后，云端服务器11的更新模块更新地图数据，将第k个目标作为地图元素作为新增元素添加到电子地图中，形成新的电子地图。云端服务器11可以将更新的电子地图通过有线或无线网络通信、V2X等技术进行传输，发放给车辆终端12。网络通信技术可以包括无线网络，例如wifi等方式进行传输方式。
在本实施例中，对不同来源的车辆上报结果执行基于最小错误率的贝叶斯估计，对目标存在状态下的后验概率和目标不存在状态下的后验概率进行比较，进而判定决策。可最大程度减少云端数据融合的错误率，提高数据融合的精度。
进一步的，当所述第一后验概率不大于所述第二后验概率时，判所述第k个目标在所述自然环境中不存在；将第k个目标作为删除元素更新地图数据。
其中，云端服务器11的判断模块对该第一后验概率和第二后验概率进行判定， 如果第一后验概率不大于第二后验概率,判定第k个目标不存在。第一后验概率是w1状态下的后验概率，由于w 2状态表征目标不存在，第一后验概率不大于第二后验概率说明该第k个目标在自然环境中不存在的概率大，即判定第k个目标在所述自然环境中不存在。
可选的，当判定第k个目标在自然环境中不存在后，云端服务器电子地图中查找该目标，若原电子地图中不存在该元素，云端服务器11不对地图数据进行更新。
可选的，云端服务器判定第k个目标在自然环境中不存在后，若原电子地图中存在该元素，将第k个目标作为删除元素，在电子地图中删除该该元素，形成新的电子地图。云端服务器11可以将更新的电子地图通过有线或无线网络通信、V2X等技术进行传输，发放给车辆终端12。网络通信技术可以包括无线网络，例如wifi等方式进行传输方式。
在上述技术方案中，通过对目标不存在的情况的分条件进行进一步判定，可以得到更加准确的地图数据，提高了地图数据的正确度。
进一步的，第一先验概率和第二先验概率与以下因素有关：目标类型、目标的位置、时间、传感器类型、设计运行条件(ODD)类型、光照条件、距离、或天气。例如：概率值与目标、目标所在的位置有关。例如，当目标为红绿灯时，那么在路口区域出现红绿灯的先验概率高。又如，先验概率与时间有关，目标还是红绿灯，当路口修好后，将设平均3天后安装红绿灯，那么第一天红绿灯存在的概率小于低于第三天红绿灯存在的概率。
对于不同的影响因素，可以形成先验概率模板并进行先验概率量化，具体方法将在下文详细描述，此处不再赘述。
在上述技术方案中，综合考虑了多种影响因素，对于不同的目标类型考虑不同的影响目标是否存在的因素，从而降低了融合的错误率。
本申请实施例还提供了一种地图数据的更新装置800，如图12所述，该装置具体包括以下模块：
确定模块801，用于确定第k个目标信息的第一初始概率和第二初始概率，其中，所述第一初始概率P(w 1)＝a，w 1表征目标存在，所述第二初始概率P(w 2)＝1-a，w 2表征目标不存在，k为正整数，所述目标信息是车辆感知周边环境所检测到的信息；
接收模块802，用于接收所述N个车辆中的第i个车辆针对所述第k个目标发送的第一先验概率和第二先验概率，其中，所述第一先验概率为w 1状态下的概率p i,k(x i|w 1)，所述第二先验概率为w2状态下的概率p i,k(x i|w 2)，1≤i≤N，N≥2且为整数；
计算模块803，用于基于所述第一初始概率、所述第二初始概率、所述第一先验概率和所述第二先验概率，计算所述第k个目标的第一后验概率和第二后验概率，其中第一后验概率是w 1状态下的后验概率，第二后验概率是w2状态下的后验概率；
判定模块804，用于当所述第一后验概率大于所述第二后验概率时,判定所述第k个目标在所述自然环境中存在。
可选的，所述判定模块还用于：当所述第一后验概率不大于所述第二后验概率时， 判所述第k个目标在所述自然环境中不存在。
更新模块805，具体用于当地图不具有表征所述第k目标的元素，不对地图数据进行更新。
可选的，当地图具有表征所述第k目标的元素，将第k个目标作为删除元素更新地图数据。
进一步的，所述计算模块计算所述第k个目标的第一后验概率和第二后验概率，具体包括如下公式1和2：
Figure PCTCN2021117860-appb-000005
Figure PCTCN2021117860-appb-000006
其中，
Figure PCTCN2021117860-appb-000007
为第k个目标信息的第一后验概率，x i k表示第i个车辆观测到第k个目标信息，
p i，k(x i|w 1)为第i个车辆发送的针对第k个目标信息的第一先验概率，P(w1)为第一初始概率,
Figure PCTCN2021117860-appb-000008
为第k个目标信息的第二后验概率，p i，k(x i|w 2)为第i个车辆发送的针对第k个目标信息的第一先验概率，P(w2)为第二初始概率。
进一步的，所述第一初始概率和第二初始概率为预设值。
进一步的，所述第一先验概率和第二先验概率与以下因素有关：目标类型、目标的位置、时间、传感器类型、设计运行条件(ODD)类型、光照条件、距离、或天气。
需要说明的是，上述地图数据的更新装置与执行上述更新方法实施例的技术方案，其实现原理和技术效果类似，此处不再赘述。
本申请实施例还提供了一种先验概率模板生成方法，用于生成一种先验概率的模板，车辆终端12上报的第一先验概率和第二先验概率可以从该模板进行取值。
该先验概率模板生成方法可以采用统计分析的方法生成，或者也可以采用理论建模的方法。
该统计分析的方法，通过对采集的实验数据数据进行统计分析，例如，采用蒙特卡洛分析法等。
理论建模的方法，首先假设先验概率的分布模型，如高斯分布、阿尔法分布、贝塔分布，得到理论概率密度模型，再通过实验值对模型进行修正。
进一步的，所述实验值可以是车辆终端在行进的过程中或者停靠的时候，通过传感器，如激光雷达、摄像头毫米波雷达、超声波或组合惯导等传感器观测到目标信息， 例如，目标信息为红绿灯信息。车辆终端可以确定该目标信息的存在概率。存在概率表示车辆终端观测到的某个目标存在的概率。
先验概率形成方法与以下因素有关：目标类型、目标的位置、时间、传感器类型、设计运行条件(ODD)类型、光照条件、距离、或天气。
例如，当目标为红绿灯时，红绿灯所在的位置将影响先验概率的大小，在路口区域，红绿灯新增的先验概率高。又如，路口修改后，假设平均3天后安装红绿灯，则第一天红绿灯的存在概率小于第三天红绿灯的存在概率。可见，先验概率的形成方法与目标类型、目标位置、时间有关。
可选的，对于不同的传感器类型，影响先验概率的具体因素并不相同。比如，对于摄像头采集的目标信息，影响先验概率的影响因素包括：不同目标、不同光照、不同纵向距离(到目标物)、不同横向距离(到目标物)、不同天气(雾霾、晴朗、阴天)。又比如，对于激光雷达(LiDAR)采集的目标信息，影响先验概率的影响因素包括：不同目标、不同目标的反射率、不同纵向距离(到目标物)、不同横向距离(到目标物)、不同天气(雾霾、晴朗、阴天)。
可选的，先验概率模板还可以包括传感器采集的目标信息融合之后的先验概率。
可选的，还可以对先验概率进行量化，形成量化的先验概率模板。比如，不同的设计运行条件(ODD)类型，量化的先验概率模板可以如下表1所示，其中，表1中的a、b为(0，1)之间的一个实数值，a<b。
表1
索引 先验概率 备注
1 <＝a ODD条件1
2 (a,b] ODD条件2
3 >b ODD条件3
又比如，不同光照条件下，量化的先验概率模板可以如下表2所示，其中下表中的a、b为(0，1)之间的一个实数值，a<b。
表2
索引 先验概率 备注
1 <＝0.3 光照条件1
2 (0.3,0.5] 光照条件2
3 >0.5 光照条件3
再比如，不同距离条件下，量化先验概率模板可以如下表3所示，其中，下表中的a、b为(0，1)之间的一个实数值，a<b。
表3
索引 先验概率 备注
1 <＝0.4 距离条件1
2 (0.4,0.6] 距离条件2
3 >0.6 距离条件3
再比如，不同天气条件下，量化先验概率模板可以如下表4所示，其中，下表中的a、b为(0，1)之间的一个实数值，a<b。
索引 先验概率 备注
1 <＝0.4 天气条件1
2 (0.4,0.6] 天气条件2
3 >0.6 天气条件3
通过设置先验概率模板，车辆终端向云端发送第一先验概率和第二先验概率，云端利用该数据信息，进行第一后验概率和第二后验概率计算，进而可以根据第一后验概率和第二后验概率来判决是否使用地图更新数据，从而提高了发送给车辆终端的地图更新数据的使用可靠度，保障车辆终端使用地图更新数据的安全性。
本申请实施例提供的一种地图数据的更新装置的结构示意图，可以参考图5，该地图数据的更新装置包括：
存储器740和处理器730；
存储器740，用于存储计算机程序；
处理器730，用于执行存储器存储的计算机程序，以实现上述实施例中的数据传输方法、数据处理方法以及地图数据的更新方法。具体可以参见前述方法实施例中的相关描述。
可选地，存储器740既可以是独立的，也可以跟处理器730集成在一起。
当存储器740是独立于处理器730之外的器件时，地图数据的更新装置还可以包括：
总线，用于连接存储器730和处理器740。
可选地，本实施例还包括：通信接口，该通信接口可以通过总线与处理器740连接。处理器730可以控制通信接口来实现地图数据的更新装置的上述的接收和发送的功能。实现原理和技术效果与方法实施例类似，其中各个模块的功能可以参考方法实施例中相应的描述，此处不再赘述。
本申请实施例提供了一种服务器，如图5所示，包括上述实施例的传输装置、处理装置、以及更新装置。实现原理和技术效果与方法实施例类似，其中各个模块的功能可以参考方法实施例中相应的描述，此处不再赘述。
本申请实施例还提供了一种服务器，可以参考图5，该装置可以用于执行上述方法实施例中对应的各个步骤和/或流程。实现原理和技术效果类似，其中各个模块的功能可以参考方法实施例中相应的描述，此处不再赘述。
本申请实施例还提供了一种可读存储介质，包括：所述可读存储介质中存储有执行指令，当地图数据的更新装置、数据传输装置、或数据处理装置的至少一个处理器执行该执行指令时，所述数据处理装置、数据传输装置、或地图数据的更新装置执行上述实施例所述的地图数据的更新方法。实现原理和技术效果类似，其中各个模块的功能可以参考方法实施例中相应的描述，此处不再赘述。
本申请实施例还提供了一种芯片，包括至少一个处理器，所述处理器与存储器耦合，所述处理器用于读取存储器中的指令并根据所述指令执行上述方法实施例所述的方法。实现原理和技术效果类似，其中各个模块的功能可以参考方法实施例中相应的描述，此处不再赘述。
本申请实施例还提供了一种路侧单元，包括上述实施例的更新装置。实现原理和技术效果与方法实施例类似，其中各个模块的功能可以参考方法实施例中相应的描述，此处不再赘述。
本身亲该实施例还提供一种计算机指令，所公开的方法可以实施为以机器可读格式被编码在计算机可读存储介质上的或者被编码在其它非瞬时性介质或者制品上的计算机程序指令。图13示意性地示出根据这里展示的至少一些实施例而布置的示例计算机程序产品的概念性局部视图，所述示例计算机程序产品包括用于在计算设备上执行计算机进程的计算机程序。在一个实施例中，示例计算机程序产品600是使用信号承载介质601来提供的。所述信号承载介质601可以包括一个或多个程序指令602，其当被一个或多个处理器运行时可以提供以上针对图7描述的功能或者部分功能。因此，例如，参考图7中所示的实施例，步骤S01-S08的一个或多个特征可以由与信号承载介质601相关联的一个或多个指令来承担。此外，图13中的程序指令602也描述示例指令。
在一些示例中，信号承载介质601可以包含计算机可读介质603,诸如但不限于，硬盘驱动器、紧密盘(CD)、数字视频光盘(DVD)、数字磁带、存储器、只读存储记忆体(Read-Only Memory，ROM)或随机存储记忆体(Random Access Memory，RAM)等等。在一些实施方式中，信号承载介质601可以包含计算机可记录介质604,诸如但不限于，存储器、读/写(R/W)CD、R/W DVD、等等。在一些实施方式中，信号承载介 质601可以包含通信介质605,诸如但不限于，数字和/或模拟通信介质(例如，光纤电缆、波导、有线通信链路、无线通信链路、等等)。因此，例如，信号承载介质601可以由无线形式的通信介质605(例如，遵守IEEE 802.11标准或者其它传输协议的无线通信介质)来传达。一个或多个程序指令602可以是，例如，计算机可执行指令或者逻辑实施指令。在一些示例中，诸如针对图3-图5描述的计算设备的计算设备可以被配置为，响应于通过计算机可读介质603、计算机可记录介质604、和/或通信介质605中的一个或多个传达到计算设备的程序指令602，提供各种操作、功能、或者动作。应该理解，这里描述的布置仅仅是用于示例的目的。因而，本领域技术人员将理解，其它布置和其它元素(例如，机器、接口、功能、顺序、和功能组等等)能够被取而代之地使用，并且一些元素可以根据所期望的结果而一并省略。另外，所描述的元素中的许多是可以被实现为离散的或者分布式的组件的、或者以任何适当的组合和位置来结合其它组件实施的功能实体。
需要说明的是，本申请实施例中对模块的划分是示意性的，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式。在本申请的实施例中的各功能模块可以集成在一个处理模块中，也可以是各个模块单独物理存在，也可以两个或两个以上模块集成在一个模块中。上述集成的模块既可以采用硬件的形式实现，也可以采用软件功能模块的形式实现。
所述集成的模块如果以软件功能模块的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读存储介质中。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的全部或部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储接种中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)或处理器(processor)执行本申请各个实施例所述方法的全部或者部分步骤。而前述的存储介质包括：U盘、移动硬盘、ROM、RAM、磁碟或者光盘等各种可以存储程序的介质。
在上述实施例中，可以全部或者部分地通过软件、硬件、固件或者其任意组合来实现。当使用软件实现时，可以全部或部分地以计算机程序产品的形式实现。所述计算机程序产品包括一个或多个计算机指令。在计算机上加载和执行所述计算机程序指令时，全部或部分地产生按照本申请实施例所述的流程或功能。所述计算机可以是通用计算机、专用计算机、计算机网络、或者其他可编程装置。所述计算机指令可以存储在计算机可读存储介质中，或者从一个计算机可读存储介质向另一个计算机可读存储介质传输，例如，所述计算机指令可以从一个网站站点、计算机、服务器或者数据中心通过有线(例如同轴电缆、光纤、数字用户线(DSL))或无线(例如红外、无线、微波等)方式向另一个网站站点、计算机、服务器或数据中心进行传输。所述计算机可读存储介质可以是计算机能够存取的任何可用戒指或者是包含一个或多个可用介质集成的服务器、数据中心等数据存储设备。所述可用介质可以是磁性介质，(例如，软盘、硬盘、磁带)、光介质(例如，DVD)、或者半导体介质(例如固态硬盘solid State Disk(SSD)等。
所述程序指令可以以软件功能单元的形式实现并能够作为独立的产品销售或使用，所述存储器可以是任意形式的计算机可读取存储介质。基于这样的理解，本申请的技 术方案全部或部分可以以软件产品的形式体现出来，包括若干指令用以使得亿台计算机设备，具体可以是处理器，来执行本申请各个实施例中目标检测装置的全部或部分步骤。而前述的计算机可读存储介质包括：U盘、移动硬盘、只读存储器(Read-Only Memory，ROM)、随机存储存储器(Random Access Memory，RAM)、磁碟或者光盘等各种可以存储程序的介质。
本实施例以上所述的电子设备，可以用于执行上述各方法实施例的技术方案，其实现原理和技术效果类似，其中各个器件的功能可以参考实施例中相应的描述，此处不再赘述。
最后应说明的是：以上所述，仅为本申请的具体实施方式，但本申请的保护范围并不局限于此，任何在本申请揭露的技术范围内的变化或替换，都应涵盖在本申请的保护范围之内。因此，本申请的保护范围应以所述权利要求的保护范围为准。

Claims (36)

  1. 一种数据传输方法，其特征在于，所述方法包括：
    确定目标信息对应的地图元素的至少一个先验概率，其中，所述目标信息是车辆感知周边环境所检测到的关于目标的信息，所述先验概率是所述目标在第一条件下存在的概率，所述至少一个先验概率中的每个先验概率对应一个所述第一条件；
    发送所述目标信息、所述至少一个先验概率以及与所述每个先验概率对应的第一条件。
  2. 根据权利要求1所述的传输方法，其特征在于：
    所述第一条件包括所述目标信息对应的地图元素的类型、地图元素的位置、地图元素的检测时间、传感器类型、或设计运行条件中的至少一个。
  3. 根据权利要求2所述的传输方法，其特征在于：
    所述传感器类型包括激光雷达、摄像头、毫米波雷达、或超声波中的至少一个。
  4. 根据权利要求1-3中任一项所述的传输方法，其特征在于：
    所述对应关系包括表格或公式。
  5. 根据权利要求1至4中任一项所述的传输方法，其特征在于：
    不同的所述传感器类型对应的所述第一条件不同。
  6. 根据权利要求2至5中任一项所述的传输方法，其特征在于：
    当所述传感器类型为摄像头时，所述第一条件包括所述目标信息对应的光照条件、所述目标信息对应的纵向距离、所述目标信息对应的横向距离、或所述目标信息对应的天气信息中的至少一个。
  7. 根据权利要求2至5中任一项所述的传输方法，其特征在于：
    当所述传感器类型为激光雷达时，所述先验概率对应的第一条件包括所述目标信息对应的反射率、所述目标信息对应的纵向距离、所述目标信息对应的横向距离、或所述目标信息对应的天气信息中的至少一个。
  8. 根据权利要求2至5中任一项所述的传输方法，其特征在于：
    当所述传感器类型为毫米波雷达时，所述先验概率对应的第一条件包括所述目标信息对应的干扰级别、所述目标信息对应的纵向距离、或所述目标信息对应的横向距离中的至少一个。
  9. 根据权利要求1至8中任一项所述的方法，其特征在于：
    所述先验概率为贝叶斯估计算法中的先验概率的量化值。
  10. 一种数据传输装置，其特征在于，所述装置包括：
    确定模块，用于确定目标信息对应的地图元素的至少一个先验概率，其中，所述目标信息是车辆感知周边环境所检测到的关于目标的信息，所述先验概率是所述目标在第一条件下存在的概率，所述至少一个先验概率中的每个先验概率对应一个所述第一条件；
    发送模块，用于发送所述目标信息、所述至少一个先验概率以及与所述每个先验概率对应的第一条件。
  11. 根据权利要求10所述的传输装置，其特征在于：
    所述第一条件包括所述目标信息对应的地图元素的类型、地图元素的位置、地图 元素的检测时间、传感器类型、或设计运行条件中的至少一个。
  12. 根据权利要求11所述的传输装置，其特征在于：
    所述传感器类型包括激光雷达、摄像头、毫米波雷达、或超声波中的至少一个。
  13. 根据权利要求10至12任一项所述的传输装置，其特征在于：
    所述对应关系包括表格或公式。
  14. 根据权利要求11至13中任一项所述的传输装置，其特征在于：
    不同的所述传感器类型对应的所述第一条件不同。
  15. 根据权利要求11至14中任一项所述的传输装置，其特征在于：
    当所述传感器类型为摄像头时，所述第一条件包括所述目标信息对应的光照条件、所述目标信息对应的纵向距离、所述目标信息对应的横向距离、或所述目标信息对应的天气信息中的至少一个。
  16. 根据权利要求11至14中任一项所述的传输装置，其特征在于：
    当所述传感器类型为激光雷达时，所述先验概率对应的第一条件包括所述目标信息对应的反射率、所述目标信息对应的纵向距离、所述目标信息对应的横向距离、或所述目标信息对应的天气信息中的至少一个。
  17. 根据权利要求11至14中任一项所述的传输装置，其特征在于：
    当所述传感器类型为毫米波雷达时，所述先验概率对应的第一条件包括所述目标信息对应的干扰级别、所述目标信息对应的纵向距离、或所述目标信息对应的横向距离中的至少一个。
  18. 根据权利要求10至17中任一项所述的传输装置，其特征在于：
    所述先验概率为贝叶斯估计算法中的先验概率的量化值。
  19. 一种数据处理方法，其特征在于，所述方法包括：
    接收目标信息、至少一个先验概率以及与所述至少一个先验概率中的每个先验概率对应的第一条件，其中，所述目标信息为车辆感知周边环境所检测到的关于目标的信息，所述先验概率是所述目标在第一条件下存在的概率，所述至少一个先验概率中的每个先验概率对应一个所述第一条件；
    基于所述至少一个先验概率计算所述目标信息的后验概率；
    基于所述后验概率,判定所述目标在所述自然环境中是否存在。
  20. 根据权利要求19所述的数据处理方法，其特征在于，所述接收至少一个先验概率包括：
    接收第一先验概率和第二先验概率，其中，所述第一先验概率是在w1状态下的先验概率，所述w1表征所述目标存在，所述第二先验概率是w2状态下的先验概率，所述w2表征所述目标不存在。
  21. 根据权利要求20所述的方法，其特征在于，所述基于所述至少一个先验概率计算所述目标信息的后验概率包括：
    基于所述第一先验概率计算第一后验概率；
    基于所述第二先验概率计算第二后验概率；
    其中第一后验概率是所述w1状态下的后验概率，第二后验概率是所述w2状态下 的后验概率。
  22. 根据权利要求21所述的方法，其特征在于，所述基于所述后验概率,判定所述目标在所述自然环境中是否存在包括：
    当所述第一后验概率大于所述第二后验概率时，判定所述目标在所述自然环境中存在；或者
    当所述第一后验概率不大于所述第二后验概率时，判定所述目标在所述自然环境中不存在。
  23. 根据权利要求19至22中任一项所述的方法，其特征在于：
    所述先验概率为贝叶斯估计算法中的先验概率的量化值；
    或者
    所述后验概率为贝叶斯估计算法中的后验概率的量化值。
  24. 一种数据处理装置，其特征在于，所述装置包括：
    接收模块，用于接收目标信息、至少一个先验概率以及与所述至少一个先验概率中的每个先验概率对应的第一条件，其中，所述目标信息反映了车辆感知周边环境所检测到的关于目标的信息，所述先验概率是所述目标在第一条件下存在的概率，所述至少一个先验概率中的每个先验概率对应一个所述第一条件；
    计算模块，用于基于所述至少一个先验概率计算所述目标信息的后验概率；
    判定模块，用于基于所述后验概率,判定所述目标在所述自然环境中是否存在。
  25. 根据权利要求24所述的装置，其特征在于，所述接收模块用于：
    接收第一先验概率和第二先验概率，其中，所述第一先验概率是在w1状态下的先验概率，所述w1表征所述目标存在，所述第二先验概率是w2状态下的先验概率，所述w2表征所述目标不存在。
  26. 根据权利要求25所述的装置，其特征在于，所述计算模块具体用于：
    基于所述第一先验概率计算第一后验概率；
    基于所述第二先验概率计算第二后验概率；
    其中第一后验概率是所述w1状态下的后验概率，第二后验概率是所述w2状态下的后验概率。
  27. 根据权利要求24至26中任一项所述的装置，其特征在于，所述判定模块具体用于：
    当所述第一后验概率大于所述第二后验概率时，判定所述目标在所述自然环境中存在；或者
    当所述第一后验概率不大于所述第二后验概率时，判定所述目标在所述自然环境中不存在。
  28. 根据权利要求24至27中任一项所述的装置，其特征在于：
    所述先验概率为贝叶斯估计算法中的先验概率的量化值；
    或者
    所述后验概率为贝叶斯估计算法中的后验概率的量化值。
  29. 一种数据传输装置，其特征在于，包括：
    存储器，用于存储程序指令；
    处理器，用于当调用并执行存储器中的程序指令时，执行如权利要求1-9任意一项所述的数据传输方法。
  30. 一种可读存储介质，其特征在于，所述可读存储介质中存储有执行指令，当数据传输装置的至少一个处理器执行该执行指令时，所述数据传输装置执行如权利要求1-9任意一项所述的数据传输方法。
  31. 一种数据处理装置，其特征在于，包括：
    存储器，用于存储程序指令；
    处理器，用于当调用并执行存储器中的程序指令时，执行如权利要求19-23任一项所述的数据传输方法。
  32. 一种可读存储介质，其特征在于，所述可读存储介质存储有执行指令，当数据处理装置的至少一个处理器执行该执行指令时，所述数据处理装置执行如权利要求19-23任一项所述的数据传输方法。
  33. 一种服务器，其特征在于，包括如权利要求24-28任一项所述的数据处理装置，或者包括如权利要求31所述的数据处理装置。
  34. 一种芯片，其特征在于，包括至少一个处理器，所述处理器与存储器耦合，所述处理器用于读取存储器中的指令并根据所述指令执行如权利要求1-9或权利要求19-23中任一项所述的方法。
  35. 一种计算机程序产品，其特征在于，当其在计算机上运行时，使得计算机执行如权利要求1-9或权利要求19-23中任一项任一项所述的方法。
  36. 一种路侧单元，其特征在于，包括如权利要求10-18或权利要求29任一项所述的数据传输装置，或者包括如权利要求24-28或权利要求31任一项所述的数据处理装置。
PCT/CN2021/117860 2020-09-29 2021-09-11 一种地图数据的传输方法及装置 WO2022068558A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202011054715.7 2020-09-29
CN202011054715.7A CN114363308B (zh) 2020-09-29 2020-09-29 一种地图数据的传输方法及装置

Publications (1)

Publication Number Publication Date
WO2022068558A1 true WO2022068558A1 (zh) 2022-04-07

Family

ID=80949633

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/117860 WO2022068558A1 (zh) 2020-09-29 2021-09-11 一种地图数据的传输方法及装置

Country Status (2)

Country Link
CN (1) CN114363308B (zh)
WO (1) WO2022068558A1 (zh)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107561549A (zh) * 2017-08-17 2018-01-09 广州视源电子科技股份有限公司 终端位置的重定位方法、装置、终端及存储介质
US20180047174A1 (en) * 2016-08-10 2018-02-15 Mitsubishi Heavy Industries, Ltd. Target monitoring system and target monitoring method
US20190234751A1 (en) * 2018-02-01 2019-08-01 Didi Research America, Llc Probabilistic navigation system and method
US20200207369A1 (en) * 2018-12-26 2020-07-02 Uatc, Llc All Mover Priors

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7860344B1 (en) * 2005-05-06 2010-12-28 Stochastech Corporation Tracking apparatus and methods using image processing noise reduction
US9114893B2 (en) * 2012-07-26 2015-08-25 Mitsubishi Electric Research Laboratories, Inc. System and method for estimating states of spacecraft in planet-moon environment
CN105606102B (zh) * 2016-01-27 2017-04-19 中国地质大学（武汉） 一种基于格网模型的pdr室内定位方法及系统
CN107422326B (zh) * 2017-03-30 2020-09-11 燕山大学 基于贝叶斯估计的水下目标追踪方法
CN110146909B (zh) * 2018-09-06 2022-03-15 腾讯科技（深圳）有限公司 一种定位数据处理方法
CN111476286A (zh) * 2020-04-02 2020-07-31 哈尔滨工程大学 一种移动机器人构建地图方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20180047174A1 (en) * 2016-08-10 2018-02-15 Mitsubishi Heavy Industries, Ltd. Target monitoring system and target monitoring method
CN107561549A (zh) * 2017-08-17 2018-01-09 广州视源电子科技股份有限公司 终端位置的重定位方法、装置、终端及存储介质
US20190234751A1 (en) * 2018-02-01 2019-08-01 Didi Research America, Llc Probabilistic navigation system and method
US20200207369A1 (en) * 2018-12-26 2020-07-02 Uatc, Llc All Mover Priors

Also Published As

Publication number Publication date
CN114363308B (zh) 2023-06-02
CN114363308A (zh) 2022-04-15

Similar Documents

Publication Publication Date Title
EP3629059B1 (en) Sharing classified objects perceived by autonomous vehicles
US20210302585A1 (en) Smart navigation method and system based on topological map
JP2022543439A (ja) 地図更新方法、装置、および記憶媒体
WO2022142664A1 (zh) 交通信息的传输方法、装置、介质、电子设备和程序产品
EP4089659A1 (en) Map updating method, apparatus and device
US20200257931A1 (en) Field theory based perception for autonomous vehicles
US20230057394A1 (en) Cooperative vehicle-infrastructure processing method and apparatus, electronic device, and storage medium
US11408739B2 (en) Location correction utilizing vehicle communication networks
US20220343758A1 (en) Data Transmission Method and Apparatus
CN115540896A (zh) 路径规划方法、装置、电子设备和计算机可读介质
WO2023028774A1 (zh) 一种激光雷达的标定方法、装置及存储介质
US20230119940A1 (en) Map update data processing method, apparatus, and system
WO2023155580A1 (zh) 一种对象识别方法和装置
KR20230007256A (ko) 도로 데이터 융합의 지도 생성 방법, 장치 및 전자 기기
EP3800443B1 (en) Database construction method, positioning method and relevant device therefor
CN113119999B (zh) 自动驾驶特征的确定方法、装置、设备、介质及程序产品
CN117389305A (zh) 一种无人机巡检路径规划方法、系统、设备及介质
WO2022068558A1 (zh) 一种地图数据的传输方法及装置
CN114724116B (zh) 车辆通行信息生成方法、装置、设备和计算机可读介质
CN113269827B (zh) 一种实现自动化标定的方法和计算装置
Xue et al. Point-cloud map update for connected and autonomous vehicles based on vehicle infrastructure cooperation: framework and field experiments
Wang et al. Development and verification of cooperative adaptive cruise control via LTE‐V
JP2021076593A (ja) 構築及び位置特定技術分野に焦点を当てた動的マップ生成
CN112612277A (zh) 车辆轨迹规划方法、装置、设备和计算机可读介质
CN113056715A (zh) 自主运载工具的基于场论的感知

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21874219

Country of ref document: EP

Kind code of ref document: A1