WO2024099111A1 - Ai服务的处理方法及设备 - Google Patents
Ai服务的处理方法及设备 Download PDFInfo
- Publication number
- WO2024099111A1 WO2024099111A1 PCT/CN2023/127307 CN2023127307W WO2024099111A1 WO 2024099111 A1 WO2024099111 A1 WO 2024099111A1 CN 2023127307 W CN2023127307 W CN 2023127307W WO 2024099111 A1 WO2024099111 A1 WO 2024099111A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- information
- service
- requirement
- target device
- quality
- Prior art date
Links
- 238000003672 processing method Methods 0.000 title claims abstract description 24
- 238000013473 artificial intelligence Methods 0.000 claims description 288
- 238000000034 method Methods 0.000 claims description 148
- 230000006870 function Effects 0.000 claims description 112
- 238000001303 quality assessment method Methods 0.000 claims description 65
- 238000011156 evaluation Methods 0.000 claims description 62
- 230000015654 memory Effects 0.000 claims description 36
- 238000012545 processing Methods 0.000 claims description 36
- 238000012423 maintenance Methods 0.000 claims description 30
- 230000008447 perception Effects 0.000 claims description 28
- 230000000977 initiatory effect Effects 0.000 claims description 20
- 238000013439 planning Methods 0.000 claims description 19
- 238000004364 calculation method Methods 0.000 claims description 18
- 238000004422 calculation algorithm Methods 0.000 claims description 16
- 238000013441 quality evaluation Methods 0.000 claims description 15
- 230000001133 acceleration Effects 0.000 claims description 14
- 238000013507 mapping Methods 0.000 claims description 9
- 230000003247 decreasing effect Effects 0.000 claims description 8
- 230000036461 convulsion Effects 0.000 claims description 7
- 238000012805 post-processing Methods 0.000 claims description 7
- 238000007781 pre-processing Methods 0.000 claims description 7
- 238000004891 communication Methods 0.000 abstract description 35
- 230000000694 effects Effects 0.000 description 13
- 238000010586 diagram Methods 0.000 description 11
- 238000010295 mobile communication Methods 0.000 description 11
- 238000007726 management method Methods 0.000 description 10
- 238000005516 engineering process Methods 0.000 description 6
- 238000004590 computer program Methods 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 101000827703 Homo sapiens Polyphosphoinositide phosphatase Proteins 0.000 description 3
- 102100023591 Polyphosphoinositide phosphatase Human genes 0.000 description 3
- 101100012902 Saccharomyces cerevisiae (strain ATCC 204508 / S288c) FIG2 gene Proteins 0.000 description 3
- 101100233916 Saccharomyces cerevisiae (strain ATCC 204508 / S288c) KAR5 gene Proteins 0.000 description 3
- 230000003068 static effect Effects 0.000 description 3
- 101001121408 Homo sapiens L-amino-acid oxidase Proteins 0.000 description 2
- 102100026388 L-amino-acid oxidase Human genes 0.000 description 2
- 230000003190 augmentative effect Effects 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 101150074586 RAN3 gene Proteins 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000013523 data management Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000007599 discharging Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 239000004984 smart glass Substances 0.000 description 1
- 238000005406 washing Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/30—Services specially adapted for particular environments, situations or purposes
- H04W4/40—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P]
- H04W4/44—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P] for communication between vehicles and infrastructures, e.g. vehicle-to-cloud [V2C] or vehicle-to-home [V2H]
Definitions
- the present application belongs to the field of communication technology, and specifically relates to a processing method and device for AI services.
- AI artificial intelligence
- CSI channel state information
- AI-based beam management CSI-based beam management
- AI-based positioning 3GPP RAN3 discussed AI-based energy saving, AI-based load balancing, etc.
- 3GPP RAN3 discussed AI-based energy saving, AI-based load balancing, etc.
- the future mobile communication system can not only provide AI services for the internal communication system, but also provide AI services for external applications.
- AI services for external applications of the communication network such as unmanned driving, drones, and robots.
- the wireless network of the mobile communication system is only used as a communication channel, that is, it is only used to transmit information, and cannot provide AI services for external applications. Therefore, how to better support AI services in the mobile communication system is a problem that needs to be solved.
- the embodiments of the present application provide a method and device for processing AI services, which can solve the problem of how to better support AI services in a mobile communication system.
- a method for processing an AI service including:
- the network-side device receives first information from the target device, where the first information includes an AI service identifier of the first service and indication information for indicating an AI service quality requirement;
- the AI service quality requirement includes at least one of the following types of requirements: safety requirements, comfort requirements, high efficiency requirements, and delay requirements;
- the safety requirement is used to characterize the anti-collision capability of the target device;
- the comfort requirement is used to characterize the riding comfort of the target device;
- the high efficiency requirement is used to characterize the efficiency of the target device in completing a task;
- the delay requirement is used to characterize the time difference between the target device initiating an AI service request and obtaining a service result;
- the network side device feeds back the service result of the first service to the target device based on the AI service identifier and the AI service quality requirement.
- a method for processing an AI service including:
- the target device sends first information to the network side device, where the first information includes an AI service identifier of the first service and indication information for indicating AI service quality requirements;
- the AI service quality requirements include at least one of the following types of requirements: safety requirements, comfort requirements, high efficiency requirements, and delay requirements;
- the safety requirements are used to characterize the anti-collision capability of the target device;
- the comfort requirements are used to characterize the riding comfort of the target device;
- the high efficiency requirements are used to characterize the efficiency of the target device in completing tasks;
- the delay requirements are used to characterize the time difference between the target device initiating an AI service request and obtaining a service result;
- the target device receives a service result of the first service sent by the network side device.
- a processing device for an AI service including:
- a receiving module is configured to receive first information from a target device, wherein the first information includes an AI service identifier of a first service and indication information for indicating an AI service quality requirement;
- the AI service quality requirement includes the following: At least one type of requirement: safety requirement, comfort requirement, efficiency requirement and delay requirement; the safety requirement is used to characterize the anti-collision capability of the target device; the comfort requirement is used to characterize the riding comfort of the target device; the efficiency requirement is used to characterize the efficiency of the target device in completing tasks; the delay requirement is used to characterize the time difference between the target device initiating an AI service request and obtaining the service result;
- the processing module is used to feed back the service result of the first service to the target device based on the AI service identifier and the AI service quality requirement.
- a processing device for an AI service including:
- a sending module configured to send first information to a network side device, wherein the first information includes an AI service identifier of a first service and indication information for indicating an AI service quality requirement;
- the AI service quality requirement includes at least one of the following types of requirements: safety requirements, comfort requirements, high efficiency requirements, and delay requirements;
- the safety requirement is used to characterize the anti-collision capability of the target device;
- the comfort requirement is used to characterize the riding comfort of the target device;
- the high efficiency requirement is used to characterize the efficiency of the target device in completing a task;
- the delay requirement is used to characterize the time difference between the target device initiating an AI service request and obtaining a service result;
- the receiving module is used to receive the service result sent by the network side device.
- a network side device which includes a processor and a memory, wherein the memory stores programs or instructions that can be run on the processor, and when the program or instructions are executed by the processor, the steps of the method described in the first aspect are implemented.
- a network side device comprising a processor and a communication interface, wherein the communication interface is used to receive first information from a target device, the first information comprising an AI service identifier of a first service and indication information for indicating AI service quality requirements;
- the AI service quality requirements comprise at least one of the following types of requirements: safety requirements, comfort requirements, high efficiency requirements and delay requirements; the safety requirements are used to characterize the collision avoidance capability of the target device; the comfort requirements are used to characterize the riding comfort of the target device; the high efficiency requirements are used to characterize the efficiency of the target device in completing tasks; the delay requirements are used to characterize the time difference between the target device initiating an AI service request and obtaining a service result; the processor is used to feedback the service result of the first service to the target device based on the AI service identifier and the AI service quality requirements.
- a target device which includes a processor and a memory, wherein the memory stores a program or instruction that can be executed on the processor, and when the program or instruction is executed by the processor, the steps of the method described in the second aspect are implemented.
- a target device comprising a processor and a communication interface, wherein the communication interface is used to send first information to a network side device, the first information comprising an AI service identifier of a first service and indication information for indicating AI service quality requirements;
- the AI service quality requirements comprise at least one of the following types of requirements: safety requirements, comfort requirements, efficiency requirements and delay requirements; the safety requirements are used to characterize the collision avoidance capability of the target device; the comfort requirements are used to characterize the riding comfort of the target device; the efficiency requirements are used to characterize the efficiency of the target device in completing tasks; the delay requirements are used to characterize the time difference between the target device initiating an AI service request and obtaining a service result; the communication interface is also used to receive the service result of the first service sent by the network side device.
- a communication system comprising: a target device and a network side device, wherein the target device can be used to execute the steps of the processing method of the AI service as described in the second aspect, and the network side device can be used to execute the steps of the processing method of the AI service as described in the first aspect.
- a readable storage medium on which a program or instruction is stored.
- the program or instruction is executed by a processor, the steps of the method described in the first aspect are implemented, or the steps of the method described in the second aspect are implemented.
- a chip comprising a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run a program or instruction to implement the method described in the first aspect, or to implement the method described in the second aspect.
- a computer program/program product is provided, wherein the computer program/program product is stored in a storage medium, and the computer program/program product is executed by at least one processor to implement the steps of the processing method of the AI service as described in the first aspect or the second aspect.
- a network side device receives first information from a target device, the first information including an AI service identifier of a first service and indication information for indicating AI service quality requirements;
- the AI service quality requirements include at least one of the following types of requirements: safety requirements, comfort requirements, efficiency requirements, and delay requirements; safety requirements are used to characterize the anti-collision capability of the target device; comfort requirements are used to characterize the riding comfort of the target device; efficiency requirements are used to characterize the efficiency of the target device in completing tasks; delay requirements are used to characterize the time difference between the target device initiating an AI service request and obtaining a service result; further, the network side device obtains the service result of the first service based on the AI service identifier and the AI service quality requirements, and feeds back the service result to the target device, that is, the network side device can provide services to the target device based on the AI service quality requirements, can improve the AI service quality, and thus achieve better support for AI services in mobile communication systems.
- FIG1 is a structural diagram of a wireless communication system applicable to an embodiment of the present application.
- FIG2 is a flowchart of a method for processing an AI service provided in an embodiment of the present application
- FIG3 is a schematic diagram of one of the interactive flow charts of the AI service processing method provided in an embodiment of the present application.
- FIG4 is a second flow chart of the processing method of the AI service provided in the embodiment of the present application.
- FIG5 is a schematic diagram of a structure of a processing device for an AI service provided in an embodiment of the present application.
- FIG6 is a second structural diagram of a processing device for an AI service provided in an embodiment of the present application.
- FIG7 is a schematic diagram of the structure of a communication device provided in an embodiment of the present application.
- FIG8 is a schematic diagram of the structure of a terminal provided in an embodiment of the present application.
- FIG. 9 is a schematic diagram of the structure of a network side device according to an embodiment of the present application.
- first, second, etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the terms used in this way are interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by “first” and “second” are generally of the same type, and the number of objects is not limited.
- the first object can be one or more.
- “and/or” in the specification and claims represents at least one of the connected objects, and the character “/" generally represents that the objects associated with each other are in an "or” relationship.
- LTE Long Term Evolution
- LTE-A Long Term Evolution-Advanced
- CDMA Code Division Multiple Access
- TDMA Time Division Multiple Access
- FDMA Frequency Division Multiple Access
- OFDMA Orthogonal Frequency Division Multiple Access
- SC-FDMA Single-carrier Frequency Division Multiple Access
- LTE Long Term Evolution
- LTE-A Long Term Evolution-Advanced
- the techniques may be used for the systems and radio technologies mentioned above as well as for other systems and radio technologies.
- the following description describes a New Radio (NR) system for example purposes, and NR terminology is used in most of the following description, but the techniques may also be applied to applications other than NR system applications, such as a 6th Generation (6G) communication system.
- NR New Radio
- 6G 6th Generation
- FIG1 shows a block diagram of a wireless communication system applicable to an embodiment of the present application.
- the wireless communication system includes a terminal 11 and a network side device 12 .
- the terminal 11 can be a mobile phone, a tablet computer (Tablet Personal Computer), a laptop computer (Laptop Computer) or a notebook computer, a personal digital assistant (Personal Digital Assistant, PDA), a handheld computer, a netbook, an ultra-mobile personal computer (ultra-mobile personal computer, UMPC), a mobile Internet device (Mobile Internet Device, MID), augmented reality (augmented reality, AR)/virtual reality (virtual reality, VR) equipment, a robot, a wearable device (Wearable Device), a vehicle-mounted device (VUE), a pedestrian terminal (PUE), a smart home (home appliances with wireless communication functions, such as refrigerators, televisions, washing machines or furniture, etc.), a game console, a personal computer (personal computer, PC), an ATM or a self-service machine and other terminal side devices, and the wear
- the network side device 12 may include an access network device or a core network device, wherein the access network device 12 may also be referred to as a radio access network device, a radio access network (RAN), a radio access network function or a radio access network unit.
- the access network device 12 may include a base station, a WLAN access point or a WiFi node, etc.
- the base station may be referred to as a node B, an evolved node B (eNB), an access point, a base transceiver station (BTS), a radio base station, a radio transceiver, a basic service set (BSS), an extended service set (ESS), a home B node, a home evolved B node, a transmitting and receiving point (TRP) or other appropriate terms in the field, as long as the same technical effect is achieved, the base station is not limited to a specific technical vocabulary, it should be noted that in the embodiment of the present application, only the base station in the NR system is used as an example for introduction, and the specific type of the base station is not limited.
- the core network equipment may include but is not limited to at least one of the following: core network nodes, core network functions, mobility management entity (Mobility Management Entity, MME), access mobility management function (Access and Mobility Management Function, AMF), session management function (Session Management Function, SMF), user plane function (User Plane Function, UPF), policy control function (Policy Control Function, PCF), policy and charging rules function unit (Policy and Charging Rules Function, PCRF), edge application service discovery function (Edge Application Server Discovery ...
- MME mobility management entity
- AMF Access and Mobility Management Function
- SMF Session Management Function
- SMF Session Management Function
- UPF User Plane Function
- Policy Control Function Policy Control Function
- PCRF Policy and Charging Rules Function
- edge application service discovery function Edge Application Server Discovery ...
- mobile communication systems can not only provide AI services to the communication system internally, but also to external applications.
- AI services can be provided to external applications of communication networks such as unmanned driving, drones, and robots. These external applications in various industries have different service quality levels. If the service quality indicators in the relevant communication system are used as service indicators for external services, the service quality requirements of external applications cannot be matched. Therefore, an embodiment of the present application provides a method for processing AI services, which carries a message for indicating AI when sending an AI service request. The indication information of the service quality requirement enables the mobile communication system to receive the AI service quality requirement, so that the mobile communication system can provide higher quality services to external applications based on the AI service quality requirement.
- FIG2 is a flowchart of a method for processing an AI service provided in an embodiment of the present application. As shown in FIG2 , the method provided in this embodiment includes:
- Step 101 The network-side device receives first information from the target device, where the first information includes an AI service identifier of the first service and indication information for indicating AI service quality requirements;
- the AI service quality requirements include at least one of the following types of requirements: safety requirements, comfort requirements, efficiency requirements, and latency requirements;
- the safety requirements are used to characterize the anti-collision capability of the target device;
- the comfort requirements are used to characterize the riding comfort of the target device;
- the efficiency requirements are used to characterize the efficiency of the target device in completing tasks;
- the latency requirements are used to characterize the time difference between the target device initiating an AI service request and obtaining a service result;
- network side devices include base station functions or core network functions
- target devices include terminals and vehicles.
- the target device sends first information to the network side device.
- the first information may be an AI service request.
- the network side device receives the first information.
- the first information includes: an AI service identifier and indication information for indicating AI service quality requirements.
- the service requested by the target device is environmental perception, such as the perception of lane lines, static objects and dynamic objects, etc., or the service requested by the target device is path planning, such as providing a trajectory including multiple trajectory point identifiers, each trajectory point identifier corresponding to a distance and speed;
- the target device is, for example, a vehicle-mounted terminal
- the safety requirement can be a cost representation calculated by the distance between the vehicle and the obstacle
- the comfort requirement can be a cost representation calculated by parameters such as the vehicle's path orientation deviation and acceleration
- the efficiency requirement is used to characterize the efficiency of the target device in completing the task, such as the cost representation calculated by parameters such as the path driving time, the deviation from the ideal obstacle-free path, and whether the road boundary is exceeded.
- the latency requirement is used to characterize the time difference between the target device initiating the AI service request and obtaining the service result fed back by the network-side device, for example, the latency requirement is less than 100ms.
- Step 102 The network-side device feeds back the service result of the first service to the target device based on the AI service identifier and the AI service quality requirement.
- the network side device provides corresponding services based on the AI service identifier and the AI service quality requirements, that is, obtains the service result of the first service corresponding to the AI service identifier, such as obtaining the service result based on the AI model.
- the output of the AI model can be iterated according to the AI service quality requirements to obtain service results with better service quality, and the service results are fed back to the target device.
- a network side device receives first information from a target device, the first information includes an AI service identifier of a first service and indication information for indicating AI service quality requirements;
- the AI service quality requirements include at least one of the following types of requirements: safety requirements, comfort requirements, high efficiency requirements, and delay requirements;
- the safety requirements are used to characterize the anti-collision capability of the target device;
- the comfort requirements are used to characterize the riding comfort of the target device;
- the high efficiency requirements are used to characterize the efficiency of the target device in completing tasks;
- the delay requirements are used to characterize the time difference between the target device initiating an AI service request and obtaining a service result;
- the network side device obtains the service result of the first service based on the AI service identifier and the AI service quality requirements, and feeds back the service result to the target device, that is, the network side device can provide services to the target device based on the AI service quality requirements, can improve the AI service quality, and thus achieve better support for AI services in mobile communication systems.
- the network side device obtains information for determining a quality assessment method; the information for determining the quality assessment method includes: an executable file and operation information corresponding to the quality assessment method; the quality assessment method is used to assess the quality of the service result.
- the operation information includes at least one of the following:
- Operating environment configuration information file name, input format, output format, information indicating preprocessing method, and information indicating postprocessing method.
- the network side device obtains information for determining a quality assessment method, including at least one of the following:
- the network side device receives information from the target device for determining a quality assessment method
- the network side device receives the third-party link from the target device, and obtains information for determining the quality assessment method based on the third-party link;
- the network side device receives the Application Programming Interface (API) from the target device and obtains information for determining the quality assessment method based on the API.
- API Application Programming Interface
- the target device can send the information used to determine the quality assessment method directly to the network side device, or send it to the network side device through a third-party link, or send it to the network side device through an API.
- the network side device directly obtains the information used to determine the quality assessment method from the received information, or obtains the information used to determine the quality assessment method based on the third-party link or API.
- the network side device may evaluate the service result based on the information used to determine the quality assessment method and obtain the evaluation result.
- the services provided can also be optimized based on the evaluation results and service quality requirements, such as optimizing the AI model.
- the first information includes a security requirement indicator threshold and first parameter information for evaluating security quality; wherein the first parameter information includes at least one of the following:
- first indication information second information, and first weighted information
- the first indication information is used to indicate at least one first target parameter item included in the security quality assessment;
- the first weighted information includes the weighted values of each of the first target parameter items;
- the second information is the function mapping information of each of the first target parameter items and the security cost, or the corresponding relationship between the first distance and the security level;
- the first target parameter item includes at least one of the following:
- the first distance is the distance between the target device and the obstacle; the first protection interval threshold is used to indicate the lower limit of the protection interval of the highest security level, and the second protection interval threshold is used to indicate the lower limit of the protection interval of the lowest security level.
- the function mapping information is used, for example, to indicate an identifier of a security cost function corresponding to the first target parameter item, or content of the security cost function.
- the security level is the highest level.
- the first distance is less than the second threshold of the protection interval, a security accident will occur and the service quality requirement will not be met.
- evaluating the security quality that is, calculating the security cost
- a larger value of the security level indicates a higher security; or a smaller value of the security level indicates a higher security.
- the security level is a monotonically decreasing function of the first distance.
- the network side device determines the first distance according to at least one of the following information
- the at least one piece of information includes: a size of the target device, a heading angle, a position of the target device, and an obstacle boundary.
- the target device is, for example, a vehicle
- the heading angle is the heading angle of the vehicle.
- the first distance is the distance between the boundary of the unmanned vehicle and the boundary of the obstacle.
- the boundary of the unmanned vehicle is represented by a three-dimensional cuboid covering the unmanned vehicle.
- the boundary of the obstacle is also represented by a three-dimensional cuboid covering the obstacle.
- the first protection interval threshold is 10 meters, which means that when the first distance is greater than or equal to 10 meters, the security cost is 0;
- the second protection interval threshold is 0.1 meters, which means that when the first distance is less than 0.1 meters, the planned trajectory is invalid.
- the safety cost is negative infinity.
- the function identifier of the safety cost is 4.
- the function exp(-d/10) is selected, where d is the first distance, to calculate the safety cost, as shown in Table 1 below:
- the first information includes a comfort requirement index threshold and second parameter information for evaluating comfort quality; wherein the second parameter information includes at least one of the following:
- the second indication information is used to indicate at least one second target parameter item included in the comfort quality evaluation, and the third information is information of the comfort cost function corresponding to each of the second target parameter items; the second weighted information includes the weighted value of each of the second target parameter items; the second target parameter item includes at least one of the following:
- Path orientation deviation path heading angle, path curvature, path curvature derivative, path curvature change rate, acceleration, and jerk.
- the second indication information is used to indicate the target parameter items included in the comfort quality assessment.
- the following parameters are considered in the comfort quality assessment: orientation deviation, heading angle, and curvature of the path.
- the first information includes information about the direction deviation cost function (e.g., direction deviation cost function identifier), information about the curvature cost function of the path (function identifier), and information about the curvature change rate cost function (function identifier).
- the cost functions of these three are agreed upon in advance.
- the weighted values of the three are also agreed upon in advance. If the weighted values of the three are different from the agreed values, they can also be carried in the first information, i.e., indicated by the first information.
- the comfort assessment considers parameters: orientation deviation of the path, curvature of the path, rate of change of curvature;
- the path includes 15 planned trajectory points.
- the orientation deviation of the path is the first-order derivative of the path.
- the orientation deviation cost is calculated by squaring the first-order derivative of the path in each time interval and then adding up the values of 15 time intervals.
- the calculation method can be predefined or included in the orientation deviation cost function information of the second information.
- the curvature of the path is the second-order derivative of the path.
- the curvature cost of the path is calculated by squaring the second-order derivative of the path in each time interval; and then adding up the values of 15 time intervals.
- the curvature rate of change is the third-order derivative of the path.
- the curvature rate of change cost is calculated by squaring the third-order derivative of the path in each time interval and then adding up the values of 15 time intervals.
- the weighted values of the orientation deviation cost, the path curvature cost, and the curvature change rate cost are w1, w2, and w3, respectively.
- the total cost of comfort is:
- the first information includes the efficiency requirement indicator threshold and third parameter information for evaluating the efficiency quality; wherein the third parameter information includes at least one of the following:
- the third indication information is used to indicate at least one third target parameter item included in the high efficiency quality assessment, and the fourth information is information of the high efficiency cost function corresponding to each of the third target parameter items; the third weighted information includes the weighted value of each of the third target parameter items; the third target parameter item includes at least one of the following:
- the third indication information is used to indicate the target parameter items included in the efficiency quality evaluation.
- the following parameters are considered in the efficiency quality evaluation: deviation from the lane centerline, path travel time, path travel distance, and penalty value for exceeding the road boundary.
- the first information includes the following items:
- Deviation cost function information from the lane centerline (the penalty function for deviation from the ideal obstacle-free path is agreed upon in advance)
- the penalty value for exceeding the road boundary is 100; (which means it is different from the pre-agreed value, or the value has not been agreed upon in advance)
- Path travel time cost function information (indicates that the calculation method has been agreed in advance)
- the weighted values of the deviation cost from the lane centerline, the penalty value for exceeding the road boundary, and the path travel time cost are w1, w2, and w3, respectively.
- this path includes 15 planned trajectory points.
- the calculation method of the deviation cost from the ideal obstacle-free path is to square the difference between the planned path and the ideal obstacle-free path in each time interval, and then add up the values of 15 time intervals.
- the calculation method can be predefined or included in the orientation deviation cost function information of the third information.
- the calculation method of the path travel time cost is 1/(1+second distance), where the second distance is the total travel distance of the planned path.
- the calculation method can be predefined or included in the direction deviation cost function information of the third information.
- the total cost of high efficiency requirement is:
- AI service quality requirements also include:
- Indication information for indicating a type of quality of service requirement, and a merging method for each type of quality of service requirement, wherein the merging method includes at least one of the following:
- Weighting function information of the parameter items included in each type of service quality requirement
- Each type of service quality requirement includes weighted values of parameter items.
- the weighting function information includes, for example: the identifier of the weighting function or the content of the weighting function, such as the weighting functions corresponding to safety requirements, comfort requirements, efficiency requirements and delay requirements, and the weighted values corresponding to multiple parameters in each type, so as to obtain the service quality requirements of each type.
- the comprehensive service quality requirement is determined by the first information, including at least one of the following methods:
- the AI service quality requirement is obtained based on the first information, and then obtained based on the AI service quality requirement.
- the comprehensive service quality requirement can be directly obtained from the first information, for example, the first information directly includes
- the AI service quality requirement may include the content of the comprehensive service quality requirement, or include the content of calculating the comprehensive service quality requirement; or, the AI service quality requirement may be first obtained from the first information, and then obtained based on the AI service quality requirement.
- the first information may include multiple service quality requirements, and the comprehensive service quality requirement may be obtained from the multiple service quality requirements.
- the first information further includes at least one of the following:
- the corresponding relationship between the comprehensive service quality level and each service quality value (the service quality value is represented by cost or delay) is shown in Table 3, for example:
- the comprehensive service quality value can be obtained by mapping the values of the service quality values.
- the quality requirement indicators involved include safety, efficiency, comfort and delay.
- the comprehensive quality value is calculated as follows:
- the AI service quality requirement is expressed by at least one of the following:
- Quality level, quality requirement index threshold includes at least one of the following: safety requirement index threshold, comfort requirement index threshold, efficiency requirement index threshold and delay requirement index threshold.
- the AI service quality requirements also include at least one of the following types of requirements:
- the speed maintenance cost is obtained based on the difference between the driving speed included in the service result and the target driving speed, and the target driving speed is determined by the second information;
- the second information includes at least one of the following: the curvature of the path, the traffic rules;
- Path monotonicity means that in a planned path, the distance between the target device and the path starting point is monotonically increasing, or the distance between the target device and the path end point is monotonically decreasing;
- the target device speed upper limit is obtained based on traffic regulations and/or target device dynamics.
- the speed maintenance cost can be obtained by the difference between the driving speed calculated by the network side device and the target driving speed. Deviation from the target driving speed will be penalized.
- the target driving speed is determined by the curvature of the path, traffic rules, etc., for example, based on road speed limits, curvature and other traffic rules.
- the above-mentioned speed maintenance cost, path monotonicity, upper limit of target device speed, etc. can be used by the network side device to provide services to the target device.
- the above-mentioned information can be used as input data of the AI model of the network side device to improve the service quality.
- the first information further includes: information used to calculate the speed maintenance cost, and the information used to calculate the speed maintenance cost includes at least one of the following:
- the deviation speed penalty function indication information is, for example, the specific content of the deviation speed penalty function, or a function identifier.
- the penalty multiple may be a weight for calculating the speed maintenance cost. For example, if the deviation speed penalty function is f, then the speed maintenance cost is f(deviation speed), or if the penalty multiple is ⁇ p , then the speed maintenance cost is ⁇ p ⁇ deviation speed, where f is, for example, a function of the difference between the driving speed included in the service result and the target driving speed.
- the first information also includes at least one of the following:
- the size of the target device is the size of the target device
- the location of the target device is the location of the target device
- the target device's perception result of the surrounding objects includes at least one of the following perception results: type, distance, speed, acceleration, boundary;
- the target device s raw perception data of surrounding objects.
- the target device is, for example, a vehicle.
- the size of the target device, the location of the target device, the original perception data, the perception results, etc. included in the first information can assist the network side device in reasoning or calculating the AI model to improve the service quality provided by the network side device.
- the method further comprises:
- the network side device determines sixth information based on the first information and the information of the quality assessment method, or based on the first information, and the sixth information includes at least one of the following: quality requirement index, quality requirement index calculation method, quality requirement index input data and format, and quality requirement index threshold.
- the quality requirement index includes, for example, at least one of the following: safety, comfort, efficiency and delay index.
- a service result can be obtained to provide a service for the target device.
- an evaluation result can also be obtained.
- the method further comprises:
- the network side device determines a first evaluation result of the service result based on the sixth information and the service result;
- the network side device sends a first evaluation result of the service result to the target device.
- the network side device can send it to the target device.
- it can also send the first evaluation result of the service result to the target device.
- the target device can filter the service results based on the first evaluation result, select the service results that meet the requirements, and perform corresponding tasks based on the service results.
- the target device may also send a second evaluation result of the service result to the network side device, and the network side device receives the second evaluation result sent by the target device, where the second evaluation result is an evaluation result of the target device evaluating the service result.
- the second evaluation result can be used to optimize the services provided by the network-side device.
- the network-side device can optimize the AI model based on the second evaluation result fed back by the target device.
- the first evaluation result or the second evaluation result includes at least one of the following:
- the AI service identifier indicates that the current service is environment-aware, and the current evaluation result is, for example, that the service level of the environment-aware service result is level 5.
- the service level in the service quality requirement is level 4. The smaller the value, the higher the level, indicating that the provided service result does not meet the service quality requirement and needs to be re-optimized. Other parameters are similar.
- step 102 may be implemented in the following manner:
- the network side device determines the algorithm information for providing the service based on the AI service identifier and the seventh information; the seventh information includes at least one of the following: AI service quality requirements and comprehensive service quality requirements;
- the network-side device feeds back the service results to the target device based on the AI service identifier and algorithm information.
- the network side device determines the algorithm information of the service based on the AI service identifier and the seventh information, executes the service algorithm (such as AI model reasoning) based on the algorithm information, obtains the service result, and feeds back the service result to the target device.
- Service results such as AI model reasoning
- the algorithm information includes a reward function for training
- the algorithm information includes an inference model.
- the method further comprises:
- the network side device determines a first evaluation result of the service result corresponding to the reasoning service based on the AI service identifier, the service result corresponding to the reasoning service, and the comprehensive service quality requirements.
- the service results corresponding to the inference service include: the speed and path of the target device within the preset time range.
- the comprehensive service quality requirements include the requirements for the comprehensive service quality level.
- the speed and the target driving speed are used to calculate the speed maintenance cost.
- the comprehensive service quality level of the service result can be determined based on the speed maintenance cost. For example, the smaller the speed maintenance cost, the higher the comprehensive service quality level. Assume that the comprehensive service quality level corresponding to the speed maintenance cost is level 3, but the requirement for the comprehensive service quality level is level 5. At this time, assuming that the larger the value of the safety level is, the higher the level is, it means that the comprehensive service quality requirements are not met.
- the target device is a vehicle
- the network-side device is a base station
- the current service is a trajectory planning service.
- the service quality requirements are pre-defined and do not require additional instructions. They can be obtained directly based on the service identifier.
- the method comprises the following steps:
- Step 1 The vehicle sends a service identifier to the base station, where the service identifier is used to indicate the path planning service.
- Step 2 The base station obtains sixth information according to the service identifier.
- the sixth information includes at least one of the following: a quality requirement index, a quality requirement index calculation method, quality requirement index input data and format, and a quality requirement index threshold.
- quality requirement indicators include safety, comfort, efficiency and latency.
- the quality requirement indicator calculation method, quality requirement indicator input data and format have been pre-defined.
- the quality requirement indicator threshold is a comprehensive service quality threshold, for example, the threshold is 10. When the threshold is less than 10, the service quality requirement is met.
- Step 3 Based on the service identifier, determine the service algorithm. Execute the service algorithm (such as AI model reasoning), obtain the service result, and then verify whether the service result meets the service requirements based on the quality requirement indicator calculation method. Or select the service result with the highest service quality.
- the service algorithm such as AI model reasoning
- Step 4 Feedback service results and evaluation results.
- the evaluation result can be a comprehensive service quality value or a grade.
- the network side device determines a comprehensive service quality requirement based on the first information, and then determines a service algorithm.
- the quality requirement indicators, quality requirement indicator calculation methods, quality requirement indicator input data and formats are all pre-defined.
- quality requirements include comfort, efficiency, safety and latency.
- the quality requirement index calculation method, quality requirement index input data and format are all pre-defined. As shown in Table 8:
- the network side device can obtain more complete and richer information, thereby improving the service quality.
- the target device reduces the complexity of its own device by enjoying the services provided by the network side device.
- the network side device can identify different AI service quality requirements and thus can provide more flexible services for the target device.
- FIG4 is a second flow chart of the processing method of the AI service provided in the embodiment of the present application. As shown in FIG3 and FIG4, the method provided in the present embodiment includes:
- Step 201 The target device sends first information to the network side device, where the first information includes an AI service identifier of a first service and indication information for indicating an AI service quality requirement;
- the AI service quality requirement includes at least one of the following types of requirements: security requirement, comfort requirement, efficiency requirement, and latency requirement;
- the safety requirement is used to characterize the anti-collision capability of the target device;
- the comfort requirement is used to characterize the riding comfort of the target device;
- the efficiency requirement is used to characterize the efficiency of the target device in completing tasks;
- the latency requirement is used to characterize the time difference between the target device initiating an AI service request and obtaining the service result;
- Step 202 The target device receives a service result of the first service sent by the network side device.
- the indication information indicates the AI service quality requirement through at least one of the following:
- Quality level, quality requirement index threshold wherein the quality requirement index threshold includes at least one of the following: safety requirement index threshold, comfort requirement index threshold, efficiency requirement index threshold and delay requirement index threshold.
- the method further comprises:
- the target device sends information for determining a quality assessment method to the network side device; the quality assessment method is used to assess the quality of a service result; the information for determining the quality assessment method includes: an executable file and operation information corresponding to the quality assessment method.
- the operation information includes at least one of the following:
- Operating environment configuration information file name, input format, output format, information indicating preprocessing method, and information indicating postprocessing method.
- the target device sends information of a quality assessment method to the network side device, including at least one of the following:
- the target device sends a third-party link to the network side device, where the third-party link is used to obtain the information used to determine the quality assessment method;
- the target device sends an application programming interface API to the network side device, where the application programming interface API is used to obtain the information used to determine the quality assessment method.
- the first information includes the security requirement indicator threshold and first parameter information for evaluating security quality; wherein the first parameter information includes at least one of the following:
- first indication information second information, and first weighted information
- the first indication information is used to indicate at least one first target parameter item included in the comfort quality evaluation; the first weighted information includes weighted values of each of the first target parameter items;
- the first target parameter item includes at least one of the following:
- the first distance is the distance between the target device and the obstacle;
- the first protection interval threshold is used to represent The first protection interval threshold is used to indicate the lower limit of the protection interval of the highest security level, and the second protection interval threshold is used to indicate the lower limit of the protection interval of the lowest security level;
- the second information is the function mapping information between each of the first target parameter items and the safety cost, or the corresponding relationship between the first distance and the safety level.
- the first distance is determined based on at least one of the following information:
- the at least one item of information includes: a size of the target device, a heading angle, a position of the target device, and an obstacle boundary.
- the first information includes the comfort requirement index threshold and second parameter information for evaluating comfort quality; wherein the second parameter information includes at least one of the following:
- the second indication information is used to indicate at least one second target parameter item included in the comfort quality evaluation, and the third information is information of the comfort cost function corresponding to each of the second target parameter items; the second weighted information includes the weighted value of each of the second target parameter items; the second target parameter item includes at least one of the following:
- Path orientation deviation path heading angle, path curvature, path curvature derivative, path curvature change rate, acceleration, and jerk.
- the first information includes the high efficiency requirement indicator threshold and third parameter information for evaluating high efficiency quality; wherein the third parameter information includes at least one of the following:
- the third indication information is used to indicate at least one third target parameter item included in the high efficiency quality assessment, and the fourth information is information of the high efficiency cost function corresponding to each of the third target parameter items; the third weighted information includes the weighted value of each of the third target parameter items; the third target parameter item includes at least one of the following:
- the AI service quality requirement also includes at least one of the following types of requirements:
- the speed maintenance cost is obtained based on the difference between the driving speed included in the service result and the target driving speed, and the target driving speed is determined by the fifth information;
- the fifth information includes at least one of the following: the curvature of the path, the traffic rules;
- Path monotonicity means that in a planned path, the distance between the target device and the path starting point is monotonically increasing, or the distance between the target device and the path end point is monotonically decreasing;
- the target device speed upper limit is obtained based on traffic regulations and/or target device dynamics.
- the first information further includes: information used to calculate the speed maintenance cost, and the information used to calculate the speed maintenance cost includes at least one of the following:
- the AI service quality requirement further includes indication information for indicating a type of service quality requirement, and a merging method for each type of service quality requirement, where the merging method includes at least one of the following:
- Weighting function information of the parameter items included in each type of service quality requirement
- Each type of service quality requirement includes weighted values of parameter items.
- the first information also includes at least one of the following:
- the size of the target device is the size of the target device
- the location of the target device is the location of the target device
- the target device's perception result of the surrounding objects includes at least one of the following perception results: type, distance, speed, acceleration, boundary;
- the target device s raw perception data of surrounding objects.
- the method further comprises:
- the target device receives a first evaluation result of the service result sent by the network side device; the first evaluation result of the service result is determined based on the sixth information and the service result; the sixth information includes at least one of the following: quality requirement index, quality requirement index calculation method, quality requirement index input data and format, quality requirement index threshold.
- the method further comprises:
- the target device sends a second evaluation result to the network side device, where the second evaluation result is an evaluation result of the target device evaluating the service result.
- the first evaluation result or the second evaluation result includes at least one of the following:
- the comprehensive service quality requirement is determined by the first information, including at least one of the following methods:
- the AI service quality requirement is obtained based on the first information, and then obtained based on the AI service quality requirement.
- the first information further includes at least one of the following:
- the processing method of the AI service provided in the embodiment of the present application can be executed by a processing device of the AI service.
- the processing method of the AI service performed by the processing device of the AI service is taken as an example to illustrate the processing device of the AI service provided in the embodiment of the present application.
- FIG5 is a schematic diagram of a structure of a processing device for an AI service provided in the present application.
- the processing device for an AI service provided in the present embodiment includes:
- the receiving module 210 is used to receive first information from a target device, wherein the first information includes a first service
- the AI service quality requirements include at least one of the following types of requirements: safety requirements, comfort requirements, high efficiency requirements and delay requirements; the safety requirements are used to characterize the anti-collision capability of the target device; the comfort requirements are used to characterize the riding comfort of the target device; the high efficiency requirements are used to characterize the efficiency of the target device in completing tasks; the delay requirements are used to characterize the time difference between the target device initiating an AI service request and obtaining a service result;
- the processing module 220 is used to feed back the service result of the first service to the target device based on the AI service identifier and the AI service quality requirement.
- the indication information indicates the AI service quality requirement through at least one of the following:
- Quality level, quality requirement index threshold wherein the quality requirement index threshold includes at least one of the following: safety requirement index threshold, comfort requirement index threshold, efficiency requirement index threshold and delay requirement index threshold.
- the receiving module 210 is further configured to:
- Acquire information for determining a quality assessment method is used to assess the quality of a service result; the information for determining the quality assessment method includes: an executable file and operation information corresponding to the quality assessment method.
- the operation information includes at least one of the following:
- Operating environment configuration information file name, input format, output format, information indicating preprocessing method, and information indicating postprocessing method.
- the receiving module 210 is specifically configured to perform at least one of the following:
- API application programming interface
- the first information includes the security requirement indicator threshold and first parameter information for evaluating security quality; wherein the first parameter information includes at least one of the following:
- first indication information second information, and first weighted information
- the first indication information is used to indicate at least one first target parameter item included in the comfort quality evaluation; the first weighted information includes weighted values of each of the first target parameter items;
- the first target parameter item includes at least one of the following:
- the first distance is the distance between the target device and the obstacle; the first protection interval threshold is used to indicate the lower limit of the protection interval of the highest security level, and the second protection interval threshold is used to indicate the lower limit of the protection interval of the lowest security level;
- the second information is the function mapping information between each of the first target parameter items and the safety cost, or the corresponding relationship between the first distance and the safety level.
- processing module 220 is further configured to:
- the at least one piece of information includes: the size of the target device, the heading angle, the location of the target device, and the edge of the obstacle boundary.
- the first information includes the comfort requirement index threshold and second parameter information for evaluating comfort quality; wherein the second parameter information includes at least one of the following:
- the second indication information is used to indicate at least one second target parameter item included in the comfort quality evaluation, and the third information is information of the comfort cost function corresponding to each of the second target parameter items; the second weighted information includes the weighted value of each of the second target parameter items; the second target parameter item includes at least one of the following:
- Path orientation deviation path heading angle, path curvature, path curvature derivative, path curvature change rate, acceleration, and jerk.
- the first information includes the high efficiency requirement indicator threshold and third parameter information for high efficiency quality evaluation; wherein the third parameter information includes at least one of the following:
- the third indication information is used to indicate at least one third target parameter item included in the high efficiency quality assessment, and the fourth information is information of the high efficiency cost function corresponding to each of the third target parameter items; the third weighted information includes the weighted value of each of the third target parameter items; the third target parameter item includes at least one of the following:
- the AI service quality requirement also includes at least one of the following types of requirements:
- the speed maintenance cost is obtained based on the difference between the driving speed included in the service result and the target driving speed, and the target driving speed is determined by the fifth information;
- the fifth information includes at least one of the following: the curvature of the path, the traffic rules;
- Path monotonicity means that in a planned path, the distance between the target device and the path starting point is monotonically increasing, or the distance between the target device and the path end point is monotonically decreasing;
- the target device speed upper limit is obtained based on traffic regulations and/or target device dynamics.
- the first information further includes: information used to calculate the speed maintenance cost, and the information used to calculate the speed maintenance cost includes at least one of the following:
- the AI service quality requirement further includes: indication information for indicating a type of service quality requirement, and a merging method for each type of service quality requirement, wherein the merging method includes at least one of the following:
- Weighting function information of the parameter items included in each type of service quality requirement
- Each type of service quality requirement includes weighted values of parameter items.
- the first information also includes at least one of the following:
- the size of the target device is the size of the target device
- the location of the target device is the location of the target device
- the target device's perception result of the surrounding objects includes at least one of the following perception results: type, distance, speed, acceleration, boundary;
- the target device s raw perception data of surrounding objects.
- processing module 220 is further configured to:
- third information is determined, and the third information includes at least one of the following: quality requirement index, quality requirement index calculation method, quality requirement index input data and format, quality requirement index threshold.
- processing module 220 is further configured to:
- the network side device sends a first evaluation result of the service result to the target device.
- the receiving module 210 is further configured to:
- a second evaluation result sent by the target device is received, where the second evaluation result is an evaluation result of the target device evaluating the service result.
- the first evaluation result or the second evaluation result includes at least one of the following:
- processing module 220 is specifically configured to:
- the fourth information includes at least one of the following: AI service quality requirements and comprehensive service quality requirements;
- the service result is fed back to the target device.
- the algorithm information includes a reward function for training
- the algorithm information includes an inference model.
- processing module 220 is further configured to:
- a first evaluation result of the service result corresponding to the reasoning service is determined.
- the comprehensive service quality requirement is determined by the first information, including at least one of the following methods:
- the AI service quality requirement is obtained based on the first information, and then obtained based on the AI service quality requirement.
- the first information further includes at least one of the following:
- the device of this embodiment can be used to execute the method of any of the embodiments in the aforementioned network side method embodiments. Its specific implementation process and technical effects are similar to those in the network side method embodiments. For details, please refer to the detailed introduction in the network side method embodiments, which will not be repeated here.
- FIG6 is a second structural diagram of the processing device of the AI service provided by the present application. As shown in FIG6 , the processing device of the AI service provided by this embodiment includes:
- the sending module 310 is used to send first information to the network side device, where the first information includes an AI service identifier of the first service and indication information for indicating AI service quality requirements;
- the AI service quality requirements include at least one of the following types of requirements: safety requirements, comfort requirements, high efficiency requirements, and delay requirements;
- the safety requirements are used to characterize the anti-collision capability of the target device;
- the comfort requirements are used to characterize the riding comfort of the target device;
- the high efficiency requirements are used to characterize the efficiency of the target device in completing the task;
- the delay requirements are used to characterize the time difference between the target device initiating an AI service request and obtaining a service result;
- the receiving module 320 is used to receive the service result of the first service sent by the network side device.
- the indication information indicates the AI service quality requirement through at least one of the following:
- Quality level, quality requirement index threshold wherein the quality requirement index threshold includes at least one of the following: safety requirement index threshold, comfort requirement index threshold, efficiency requirement index threshold and delay requirement index threshold.
- the sending module 310 is further configured to:
- Information for determining a quality assessment method is sent to the network side device; the quality assessment method is used to assess the quality of the service result; the information for determining the quality assessment method includes: an executable file and operation information corresponding to the quality assessment method.
- the operation information includes at least one of the following:
- Operating environment configuration information file name, input format, output format, information indicating preprocessing method, and information indicating postprocessing method.
- the sending module 310 is specifically configured to perform at least one of the following:
- An application programming interface API is sent to the network side device, where the application programming interface API is used to obtain the information used to determine the quality assessment method.
- the first information includes the security requirement indicator threshold and first parameter information for evaluating security quality; wherein the first parameter information includes at least one of the following:
- first indication information second information, and first weighted information
- the first indication information is used to indicate at least one first target parameter item included in the comfort quality evaluation; the first weighted information includes weighted values of each of the first target parameter items;
- the first target parameter item includes at least one of the following:
- the first distance is the distance between the target device and the obstacle; the first protection interval threshold is used to indicate the lower limit of the protection interval of the highest security level, and the second protection interval threshold is used to indicate the lower limit of the protection interval of the lowest security level;
- the second information is the function mapping information between each of the first target parameter items and the safety cost, or the corresponding relationship between the first distance and the safety level.
- the first distance is determined based on at least one of the following information:
- the at least one piece of information includes: the size of the target device, the heading angle, the location of the target device, and the edge of the obstacle boundary.
- the first information includes the comfort requirement index threshold and second parameter information for evaluating comfort quality; wherein the second parameter information includes at least one of the following:
- the second indication information is used to indicate at least one second target parameter item included in the comfort quality evaluation, and the third information is information of the comfort cost function corresponding to each of the second target parameter items; the second weighted information includes the weighted value of each of the second target parameter items; the second target parameter item includes at least one of the following:
- Path orientation deviation path heading angle, path curvature, path curvature derivative, path curvature change rate, acceleration, and jerk.
- the first information includes the high efficiency requirement indicator threshold and third parameter information for high efficiency quality evaluation; wherein the third parameter information includes at least one of the following:
- the third indication information is used to indicate at least one third target parameter item included in the high efficiency quality assessment, and the fourth information is information of the high efficiency cost function corresponding to each of the third target parameter items; the third weighted information includes the weighted value of each of the third target parameter items; the third target parameter item includes at least one of the following:
- the AI service quality requirement also includes at least one of the following types of requirements:
- the speed maintenance cost is obtained based on the difference between the driving speed included in the service result and the target driving speed, and the target driving speed is determined by the fifth information;
- the fifth information includes at least one of the following: the curvature of the path, the traffic rules;
- Path monotonicity means that in a planned path, the distance between the target device and the path starting point is monotonically increasing, or the distance between the target device and the path end point is monotonically decreasing;
- the target device speed upper limit is obtained based on traffic regulations and/or target device dynamics.
- the first information further includes: information used to calculate the speed maintenance cost, and the information used to calculate the speed maintenance cost includes at least one of the following:
- the AI service quality requirement further includes indication information for indicating a type of service quality requirement, and a merging method for each type of service quality requirement, where the merging method includes at least one of the following:
- Weighting function information of the parameter items included in each type of service quality requirement
- Each type of service quality requirement includes weighted values of parameter items.
- the first information also includes at least one of the following:
- the size of the target device is the size of the target device
- the location of the target device is the location of the target device
- the target device's perception result of the surrounding objects includes at least one of the following perception results: type, distance, speed, acceleration, boundary;
- the target device s raw perception data of surrounding objects.
- the receiving module 320 is further used for:
- the first evaluation result of the service result is determined based on the sixth information and the service result;
- the sixth information includes at least one of the following: quality requirement index, quality requirement index calculation method, quality requirement index input data and format, quality requirement index threshold.
- the sending module 310 is further configured to:
- a second evaluation result is sent to the network side device, where the second evaluation result is an evaluation result of the target device evaluating the service result.
- the first evaluation result or the second evaluation result includes at least one of the following:
- the comprehensive service quality requirement is determined by the first information, including at least one of the following methods:
- the AI service quality requirement is obtained based on the first information, and then obtained based on the AI service quality requirement.
- the first information further includes at least one of the following:
- the device of this embodiment can be used to execute the method of any of the aforementioned target device side method embodiments. Its specific implementation process and technical effects are similar to those in the target device side method embodiments. For details, please refer to the detailed introduction in the target device side method embodiments, which will not be repeated here.
- the processing device of the AI service in the embodiment of the present application may be an electronic device, such as an electronic device with an operating system, or a component in an electronic device, such as an integrated circuit or a chip.
- the electronic device may be a terminal, or may be other devices other than a terminal.
- the terminal may include but is not limited to the types of terminal 11 listed above, and other devices may be servers, network attached storage (NAS), etc., which are not specifically limited in the embodiment of the present application.
- the processing device of the AI service provided in the embodiment of the present application can implement the various processes implemented by the method embodiments of Figures 2 to 4 and achieve the same technical effect. To avoid repetition, it will not be repeated here.
- the embodiment of the present application further provides a communication device 700, including a processor 701 and a memory 702, wherein the memory 702 stores a program or instruction that can be run on the processor 701.
- the communication device 700 is a target device
- the program or instruction is executed by the processor 701 to implement the various steps of the processing method embodiment of the above-mentioned AI service, and can achieve the same technical effect.
- the communication device 700 is a network side device
- the program or instruction is executed by the processor 701 to implement the various steps of the processing method embodiment of the above-mentioned AI service, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.
- the embodiment of the present application also provides a target device, including a processor and a communication interface, the communication interface is used to send a first message to a network side device, the first message includes an AI service identifier of a first service and indication information for indicating AI service quality requirements;
- the AI service quality requirements include at least one of the following types of requirements: safety requirements, comfort requirements, high efficiency requirements and delay requirements; the safety requirements are used to characterize the anti-collision capability of the target device; the comfort requirements are used to characterize the riding comfort of the target device; the high efficiency requirements are used to characterize the efficiency of the target device in completing tasks; the delay requirements are used to characterize the time difference between the target device initiating an AI service request and obtaining a service result; the communication interface is also used to receive the service result of the first service sent by the network side device.
- FIG8 is a schematic diagram of the hardware structure of a terminal that implements the embodiment of the present application.
- the terminal 1000 includes but is not limited to: a radio frequency unit 1001, a network module 1002, an audio output unit 1003, an input unit 1004, a sensor 1005, a display unit 1006, a user input unit 1007, an interface unit 1008, a memory 1009, and at least some of the components of a processor 1010.
- the terminal 1000 may also include a power source (such as a battery) for supplying power to each component, and the power source may be logically connected to the processor 1010 through a power management system, so as to manage charging, discharging, and power consumption management through the power management system.
- a power source such as a battery
- the terminal structure shown in FIG8 does not constitute a limitation on the terminal, and the terminal may include more or fewer components than shown in the figure, or combine certain components, or arrange components differently, which will not be described in detail here.
- the input unit 1004 may include a graphics processing unit (GPU) 10041 and a microphone 10042, and the graphics processor 10041 processes the image data of the static picture or video obtained by the image capture device (such as a camera) in the video capture mode or the image capture mode.
- the display unit 1006 may include a display panel 10061, and the display panel 10061 may be configured in the form of a liquid crystal display, an organic light emitting diode, etc.
- the user input unit 1007 includes at least one of a touch panel 10071 and other input devices 10072.
- the touch panel 10071 is also called a touch screen.
- the touch panel 10071 may include two parts: a touch detection device and a touch controller.
- Other input devices 10072 may include, but are not limited to, a physical keyboard, function keys (such as a volume control key, a switch key, etc.), a trackball, a mouse, and a joystick, which will not be repeated here.
- the RF unit 1001 after receiving the downlink data from the network side device, the RF unit 1001 can transmit it to the processor 1010 for processing; in addition, the RF unit 1001 can send the uplink data to the network side device.
- the RF unit 1001 includes but is not limited to an antenna, at least one amplifier, a transceiver, a coupler, a low noise amplifier, a duplexer, etc.
- the memory 1009 can be used to store software programs or instructions and various data.
- the memory 1009 may mainly include a first storage area for storing programs or instructions and a second storage area for storing data, wherein the first storage area for programs or instructions may store an operating system, an application program or instruction required for at least one function (such as a sound playback function, an image playback function, etc.), etc.
- the memory 1009 may include a volatile memory or a non-volatile memory, or the memory 1009 may include both volatile and non-volatile memories.
- Non-volatile memory includes a high-speed random access memory and may also include a non-volatile memory, wherein the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory.
- the volatile memory may be a random access memory (RAM), a static random access memory (SRAM), a dynamic random access memory (DRAM), or a flash memory.
- the memory 1009 in the embodiment of the present application includes but is not limited to these and any other suitable types of memory such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device.
- the processor 1010 may include one or more processing units; optionally, the processor 1010 may integrate an application processor and a modem processor, wherein the application processor mainly processes operations related to an operating system, a user interface, and application programs or instructions, and the modem processor mainly processes wireless communication signals, such as a baseband processor. It is understandable that the modem processor may not be integrated into the processor 1010.
- the radio frequency unit 1001 is used to send first information to the network side device, where the first information includes an AI service identifier of the first service and indication information for indicating AI service quality requirements;
- the AI service quality requirements include at least one of the following types of requirements: safety requirements, comfort requirements, high efficiency requirements and delay requirements; the safety requirements are used to characterize the anti-collision capability of the target device; the comfort requirements are used to characterize the riding comfort of the target device; the high efficiency requirements are used to characterize the efficiency of the target device in completing the task; the delay requirements are used to characterize the time difference between the target device initiating an AI service request and obtaining a service result;
- the radio frequency unit 1001 is further configured to receive a service result of the first service sent by the network side device.
- the indication information indicates the AI service quality requirement through at least one of the following:
- Quality level, quality requirement index threshold wherein the quality requirement index threshold includes at least one of the following: safety requirement index threshold, comfort requirement index threshold, efficiency requirement index threshold and delay requirement index threshold.
- the radio frequency unit 1001 is further configured to:
- Information for determining a quality assessment method is sent to the network side device; the quality assessment method is used to assess the quality of the service result; the information for determining the quality assessment method includes: an executable file and operation information corresponding to the quality assessment method.
- the operation information includes at least one of the following:
- Operating environment configuration information file name, input format, output format, information indicating preprocessing method, and information indicating postprocessing method.
- the radio frequency unit 1001 is specifically configured to perform at least one of the following:
- An application programming interface API is sent to the network side device, where the application programming interface API is used to obtain the information used to determine the quality assessment method.
- the first information includes the security requirement indicator threshold and first parameter information for evaluating security quality; wherein the first parameter information includes at least one of the following:
- first indication information second information, and first weighted information
- the first indication information is used to indicate at least one first target parameter item included in the comfort quality evaluation; the first weighted information includes weighted values of each of the first target parameter items;
- the first target parameter item includes at least one of the following:
- the first distance is the distance between the target device and the obstacle;
- the first protection interval threshold is used to represent The first protection interval threshold is used to indicate the lower limit of the protection interval of the highest security level, and the second protection interval threshold is used to indicate the lower limit of the protection interval of the lowest security level;
- the second information is the function mapping information between each of the first target parameter items and the safety cost, or the corresponding relationship between the first distance and the safety level.
- the first distance is determined based on at least one of the following information:
- the at least one item of information includes: a size of the target device, a heading angle, a position of the target device, and an obstacle boundary.
- the first information includes the comfort requirement index threshold and second parameter information for evaluating comfort quality; wherein the second parameter information includes at least one of the following:
- the second indication information is used to indicate at least one second target parameter item included in the comfort quality evaluation, and the third information is information of the comfort cost function corresponding to each of the second target parameter items; the second weighted information includes the weighted value of each of the second target parameter items; the second target parameter item includes at least one of the following:
- Path orientation deviation path heading angle, path curvature, path curvature derivative, path curvature change rate, acceleration, and jerk.
- the first information includes the high efficiency requirement indicator threshold and third parameter information for evaluating high efficiency quality; wherein the third parameter information includes at least one of the following:
- the third indication information is used to indicate at least one third target parameter item included in the high efficiency quality assessment, and the fourth information is information of the high efficiency cost function corresponding to each of the third target parameter items; the third weighted information includes the weighted value of each of the third target parameter items; the third target parameter item includes at least one of the following:
- the AI service quality requirement also includes at least one of the following types of requirements:
- the speed maintenance cost is obtained based on the difference between the driving speed included in the service result and the target driving speed, and the target driving speed is determined by the fifth information;
- the fifth information includes at least one of the following: the curvature of the path, the traffic rules;
- Path monotonicity means that in a planned path, the distance between the target device and the path starting point is monotonically increasing, or the distance between the target device and the path end point is monotonically decreasing;
- the target device speed upper limit is obtained based on traffic regulations and/or target device dynamics.
- the first information further includes: information used to calculate the speed maintenance cost, and the information used to calculate the speed maintenance cost includes at least one of the following:
- the AI service quality requirement further includes indication information for indicating a type of service quality requirement, and a merging method for each type of service quality requirement, where the merging method includes at least one of the following:
- Weighting function information of the parameter items included in each type of service quality requirement
- Each type of service quality requirement includes weighted values of parameter items.
- the first information also includes at least one of the following:
- the size of the target device is the size of the target device
- the location of the target device is the location of the target device
- the target device's perception result of the surrounding objects includes at least one of the following perception results: type, distance, speed, acceleration, boundary;
- the target device s raw perception data of surrounding objects.
- the radio frequency unit 1001 is further used for:
- the first evaluation result of the service result is determined based on the sixth information and the service result;
- the sixth information includes at least one of the following: quality requirement index, quality requirement index calculation method, quality requirement index input data and format, quality requirement index threshold.
- the radio frequency unit 1001 is further used for:
- a second evaluation result is sent to the network side device, where the second evaluation result is an evaluation result of the target device evaluating the service result.
- the first evaluation result or the second evaluation result includes at least one of the following:
- the comprehensive service quality requirement is determined by the first information, including at least one of the following methods:
- the AI service quality requirement is obtained based on the first information, and then obtained based on the AI service quality requirement.
- the first information further includes at least one of the following:
- the embodiment of the present application also provides a network-side device, including a processor and a communication interface, the communication interface is used to receive first information from a target device, the first information includes an AI service identifier of a first service and indication information for indicating AI service quality requirements; the AI service quality requirements include at least one of the following types of requirements: safety requirements, comfort requirements, high efficiency requirements and delay requirements; the safety requirements are used to characterize the anti-collision capability of the target device; the comfort requirements are used to characterize the riding comfort of the target device; the high efficiency requirements are used to characterize the efficiency of the target device in completing tasks; the delay requirements are used to characterize the time difference between the target device initiating an AI service request and obtaining a service result; the processor is used to feedback the service result to the target device based on the AI service identifier and the AI service quality requirements.
- This network-side device embodiment is similar to the above-mentioned network-side device embodiment. Corresponding to the network side device method embodiment, each implementation process and implementation method of the above method
- the embodiment of the present application further provides a network side device.
- the network side device 800 includes: an antenna 81 , a radio frequency device 82 , a baseband device 83 , a processor 85 and a memory 85 .
- the antenna 81 is connected to the radio frequency device 82 .
- the radio frequency device 82 receives information via the antenna 81 and sends the received information to the baseband device 83 for processing.
- the baseband device 83 processes the information to be sent and sends it to the radio frequency device 82 .
- the radio frequency device 82 processes the received information and sends it out through the antenna 81 .
- the above-mentioned frequency band processing device may be located in the baseband device 83 .
- the method executed by the access network device in the above embodiment may be implemented in the baseband device 83 .
- the baseband device 83 includes a baseband processor 85 and a memory 85 .
- the baseband device 83 may include, for example, at least one baseband board, on which a plurality of chips are arranged, as shown in FIG. 9 , wherein one of the chips is, for example, a baseband processor 85, which is connected to the memory 85 through a bus interface to call a program in the memory 85 and execute the access network device operations shown in the above method embodiment.
- the access network device 800 may also include a network interface 86 for interacting with the radio frequency device 82, such as a common public radio interface (CPRI).
- a network interface 86 for interacting with the radio frequency device 82, such as a common public radio interface (CPRI).
- CPRI common public radio interface
- the access network device 800 implemented in the present application also includes: instructions or programs stored in the memory 85 and executable on the processor 85.
- the processor 85 calls the instructions or programs in the memory 85 to execute the method executed by the module shown in Figure 5 and achieves the same technical effect. To avoid repetition, it will not be repeated here.
- the embodiment of the present application also provides a readable storage medium, on which a program or instruction is stored.
- a program or instruction is stored.
- each process of the embodiment of the processing method of the above-mentioned AI service is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.
- the processor is the processor in the terminal described in the above embodiment.
- the readable storage medium includes a computer readable storage medium, such as a computer read-only memory ROM, a random access memory RAM, a magnetic disk or an optical disk.
- An embodiment of the present application further provides a chip, which includes a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the various processes of the above-mentioned AI service processing method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.
- the chip mentioned in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.
- the embodiments of the present application further provide a computer program/program product, which is stored in a storage medium, and is executed by at least one processor to implement the various processes of the above-mentioned AI service processing method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.
- An embodiment of the present application also provides a communication system, including: a target device and a network side device, wherein the target device can be used to execute the steps of the processing method of the AI service as described above, and the network side device can be used to execute the steps of the processing method of the AI service as described above.
- the technical solution of the present application can be embodied in the form of a computer software product, which is stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal (which can be a mobile phone, computer, server, air conditioner, or network equipment, etc.) to execute the methods described in each embodiment of the present application.
- a storage medium such as ROM/RAM, magnetic disk, optical disk
- a terminal which can be a mobile phone, computer, server, air conditioner, or network equipment, etc.
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
本申请公开了一种AI服务的处理方法及设备,属于通信技术领域,本申请实施例的AI服务的处理方法包括:网络侧设备接收来自目标设备的第一信息,所述第一信息包括第一服务的AI服务标识和用于指示AI服务质量要求的指示信息;AI服务质量要求包括以下至少一种类型的要求:安全性要求、舒适性要求、高效性要求和时延要求;安全性要求用于表征目标设备的防碰撞能力;所述舒适性要求用于表征目标设备的乘坐舒适度;高效性要求用于表征目标设备完成任务的效率;时延要求用于表征目标设备发起AI服务请求与获取到服务结果之间的时间差;网络侧设备基于AI服务标识和AI服务质量要求,向目标设备反馈第一服务的服务结果。
Description
相关申请的交叉引用
本申请主张在2022年11月07日提交的申请号为202211387610.2的中国专利的优先权,其全部内容通过引用包含于此。
本申请属于通信技术领域,具体涉及一种AI服务的处理方法及设备。
在移动通信系统中,现在有越来越多的用例结合人工智能(Artificial Intelligence,AI)。例如在物理层有基于AI的CSI(channel state information)反馈压缩,基于AI的波束管理,基于AI的定位。3GPP RAN3讨论了基于AI的节能,基于AI的负载均衡等。在未来会有更多的结合AI的用例在移动通信系统中出现。
未来的移动通信系统不仅可以向内为通信系统提供AI服务。也可以向外部应用提供AI服务。例如,为无人驾驶、无人机、机器人等通信网络外部应用提供AI服务。目前移动通信系统的无线网络只是作为一个通信管道,即只用于传输信息,而并不能为外部应用提供AI服务,因此如何更好地在移动通信系统中支持AI服务,是需要解决的问题。
发明内容
本申请实施例提供一种AI服务的处理方法及设备,能够解决如何更好地在移动通信系统中支持AI服务的问题。
第一方面,提供了一种AI服务的处理方法,包括:
网络侧设备接收来自目标设备的第一信息,所述第一信息包括第一服务的AI服务标识和用于指示AI服务质量要求的指示信息;所述AI服务质量要求包括以下至少一种类型的要求:安全性要求、舒适性要求、高效性要求和时延要求;所述安全性要求用于表征所述目标设备的防碰撞能力;所述舒适性要求用于表征所述目标设备的乘坐舒适度;所述高效性要求用于表征所述目标设备完成任务的效率;所述时延要求用于表征所述目标设备发起AI服务请求与获取到服务结果之间的时间差;
所述网络侧设备基于所述AI服务标识和AI服务质量要求,向目标设备反馈第一服务的服务结果。
第二方面,提供了一种AI服务的处理方法,包括:
目标设备向网络侧设备发送第一信息,所述第一信息包括第一服务的AI服务标识和用于指示AI服务质量要求的指示信息;所述AI服务质量要求包括以下至少一种类型的要求:安全性要求、舒适性要求、高效性要求和时延要求;所述安全性要求用于表征所述目标设备的防碰撞能力;所述舒适性要求用于表征所述目标设备的乘坐舒适度;所述高效性要求用于表征所述目标设备完成任务的效率;所述时延要求用于表征所述目标设备发起AI服务请求与获取到服务结果之间的时间差;
所述目标设备接收所述网络侧设备发送的第一服务的服务结果。
第三方面,提供了一种AI服务的处理装置,包括:
接收模块,用于接收来自目标设备的第一信息,所述第一信息包括第一服务的AI服务标识和用于指示AI服务质量要求的指示信息;所述AI服务质量要求包括以下至
少一种类型的要求:安全性要求、舒适性要求、高效性要求和时延要求;所述安全性要求用于表征所述目标设备的防碰撞能力;所述舒适性要求用于表征所述目标设备的乘坐舒适度;所述高效性要求用于表征所述目标设备完成任务的效率;所述时延要求用于表征所述目标设备发起AI服务请求与获取到服务结果之间的时间差;
处理模块,用于基于所述AI服务标识和AI服务质量要求,向目标设备反馈第一服务的服务结果。
第四方面,提供了一种AI服务的处理装置,包括:
发送模块,用于向网络侧设备发送第一信息,所述第一信息包括第一服务的AI服务标识和用于指示AI服务质量要求的指示信息;所述AI服务质量要求包括以下至少一种类型的要求:安全性要求、舒适性要求、高效性要求和时延要求;所述安全性要求用于表征所述目标设备的防碰撞能力;所述舒适性要求用于表征所述目标设备的乘坐舒适度;所述高效性要求用于表征所述目标设备完成任务的效率;所述时延要求用于表征所述目标设备发起AI服务请求与获取到服务结果之间的时间差;
接收模块,用于接收所述网络侧设备发送的服务结果。
第五方面,提供了一种网络侧设备,该网络侧设备包括处理器和存储器,所述存储器存储可在所述处理器上运行的程序或指令,所述程序或指令被所述处理器执行时实现如第一方面所述的方法的步骤。
第六方面,提供了一种网络侧设备,包括处理器及通信接口,其中,所述通信接口用于接收来自目标设备的第一信息,所述第一信息包括第一服务的AI服务标识和用于指示AI服务质量要求的指示信息;所述AI服务质量要求包括以下至少一种类型的要求:安全性要求、舒适性要求、高效性要求和时延要求;所述安全性要求用于表征所述目标设备的防碰撞能力;所述舒适性要求用于表征所述目标设备的乘坐舒适度;所述高效性要求用于表征所述目标设备完成任务的效率;所述时延要求用于表征所述目标设备发起AI服务请求与获取到服务结果之间的时间差;所述处理器用于基于所述AI服务标识和AI服务质量要求,向目标设备反馈第一服务的服务结果。
第七方面,提供了一种目标设备,该目标设备包括处理器和存储器,所述存储器存储可在所述处理器上运行的程序或指令,所述程序或指令被所述处理器执行时实现如第二方面所述的方法的步骤。
第八方面,提供了一种目标设备,包括处理器及通信接口,其中,所述通信接口用于向网络侧设备发送第一信息,所述第一信息包括第一服务的AI服务标识和用于指示AI服务质量要求的指示信息;所述AI服务质量要求包括以下至少一种类型的要求:安全性要求、舒适性要求、高效性要求和时延要求;所述安全性要求用于表征所述目标设备的防碰撞能力;所述舒适性要求用于表征所述目标设备的乘坐舒适度;所述高效性要求用于表征所述目标设备完成任务的效率;所述时延要求用于表征所述目标设备发起AI服务请求与获取到服务结果之间的时间差;所述通信接口还用于接收所述网络侧设备发送的第一服务的服务结果。
第九方面,提供了一种通信系统,包括:目标设备及网络侧设备,所述目标设备可用于执行如第二方面所述的AI服务的处理方法的步骤,所述网络侧设备可用于执行如第一方面所述的AI服务的处理方法的步骤。
第十方面,提供了一种可读存储介质,所述可读存储介质上存储程序或指令,所述程序或指令被处理器执行时实现如第一方面所述的方法的步骤,或者实现如第二方面所述的方法的步骤。
第十一方面,提供了一种芯片,所述芯片包括处理器和通信接口,所述通信接口和所述处理器耦合,所述处理器用于运行程序或指令,实现如第一方面所述的方法,或实现如第二方面所述的方法。
第十二方面,提供了一种计算机程序/程序产品,所述计算机程序/程序产品被存储在存储介质中,所述计算机程序/程序产品被至少一个处理器执行以实现如第一方面或第二方面所述的AI服务的处理方法的步骤。
在本申请实施例中,网络侧设备接收来自目标设备的第一信息,第一信息包括第一服务的AI服务标识和用于指示AI服务质量要求的指示信息;AI服务质量要求包括以下至少一种类型的要求:安全性要求、舒适性要求、高效性要求和时延要求;安全性要求用于表征所述目标设备的防碰撞能力;舒适性要求用于表征目标设备的乘坐舒适度;高效性要求用于表征目标设备完成任务的效率;时延要求用于表征目标设备发起AI服务请求与获取到服务结果之间的时间差;进一步地,网络侧设备基于AI服务标识和AI服务质量要求,获取第一服务的服务结果,并向目标设备反馈服务结果,即网络侧设备可以基于AI服务质量要求为目标设备提供服务,可以提升AI服务质量,从而实现了更好地在移动通信系统中支持AI服务。
图1是本申请实施例可应用的无线通信系统的结构图;
图2是本申请实施例提供的AI服务的处理方法的流程示意图之一;
图3是本申请实施例提供的AI服务的处理方法的交互流程示意图之一;
图4是本申请实施例提供的AI服务的处理方法的流程示意图之二;
图5是本申请实施例提供的AI服务的处理装置的结构示意图之一;
图6是本申请实施例提供的AI服务的处理装置的结构示意图之二;
图7是本申请实施例提供的通信设备的结构示意图;
图8是本申请实施例提供的终端的结构示意图;
图9是本申请实施例的网络侧设备的结构示意图。
下面将结合本申请实施例中的附图,对本申请实施例中的技术方案进行清楚描述,显然,所描述的实施例是本申请一部分实施例,而不是全部的实施例。基于本申请中的实施例,本领域普通技术人员所获得的所有其他实施例,都属于本申请保护的范围。
本申请的说明书和权利要求书中的术语“第一”、“第二”等是用于区别类似的对象,而不用于描述特定的顺序或先后次序。应该理解这样使用的术语在适当情况下可以互换,以便本申请的实施例能够以除了在这里图示或描述的那些以外的顺序实施,且“第一”、“第二”所区别的对象通常为一类,并不限定对象的个数,例如第一对象可以是一个,也可以是多个。此外,说明书以及权利要求中“和/或”表示所连接对象的至少其中之一,字符“/”一般表示前后关联对象是一种“或”的关系。
值得指出的是,本申请实施例所描述的技术不限于长期演进型(Long Term Evolution,LTE)/LTE的演进(LTE-Advanced,LTE-A)系统,还可用于其他无线通信系统,诸如码分多址(Code Division Multiple Access,CDMA)、时分多址(Time Division Multiple Access,TDMA)、频分多址(Frequency Division Multiple Access,FDMA)、正交频分多址(Orthogonal Frequency Division Multiple Access,OFDMA)、单载波频分多址(Single-carrier Frequency Division Multiple Access,SC-FDMA)和其他系统。本申请实施例中的术语“系统”和“网络”常被可互换地使用,所描述的技
术既可用于以上提及的系统和无线电技术,也可用于其他系统和无线电技术。以下描述出于示例目的描述了新空口(New Radio,NR)系统,并且在以下大部分描述中使用NR术语,但是这些技术也可应用于NR系统应用以外的应用,如第6代(6th Generation,6G)通信系统。
图1示出本申请实施例可应用的一种无线通信系统的框图。无线通信系统包括终端11和网络侧设备12。其中,终端11可以是手机、平板电脑(Tablet Personal Computer)、膝上型电脑(Laptop Computer)或称为笔记本电脑、个人数字助理(Personal Digital Assistant,PDA)、掌上电脑、上网本、超级移动个人计算机(ultra-mobile personal computer,UMPC)、移动上网装置(Mobile Internet Device,MID)、增强现实(augmented reality,AR)/虚拟现实(virtual reality,VR)设备、机器人、可穿戴式设备(Wearable Device)、车载设备(VUE)、行人终端(PUE)、智能家居(具有无线通信功能的家居设备,如冰箱、电视、洗衣机或者家具等)、游戏机、个人计算机(personal computer,PC)、柜员机或者自助机等终端侧设备,可穿戴式设备包括:智能手表、智能手环、智能耳机、智能眼镜、智能首饰(智能手镯、智能手链、智能戒指、智能项链、智能脚镯、智能脚链等)、智能腕带、智能服装等。需要说明的是,在本申请实施例并不限定终端11的具体类型。网络侧设备12可以包括接入网设备或核心网设备,其中,接入网设备12也可以称为无线接入网设备、无线接入网(Radio Access Network,RAN)、无线接入网功能或无线接入网单元。接入网设备12可以包括基站、WLAN接入点或WiFi节点等,基站可被称为节点B、演进节点B(eNB)、接入点、基收发机站(Base Transceiver Station,BTS)、无线电基站、无线电收发机、基本服务集(Basic Service Set,BSS)、扩展服务集(Extended Service Set,ESS)、家用B节点、家用演进型B节点、发送接收点(Transmitting Receiving Point,TRP)或所述领域中其他某个合适的术语,只要达到相同的技术效果,所述基站不限于特定技术词汇,需要说明的是,在本申请实施例中仅以NR系统中的基站为例进行介绍,并不限定基站的具体类型。核心网设备可以包含但不限于如下至少一项:核心网节点、核心网功能、移动管理实体(Mobility Management Entity,MME)、接入移动管理功能(Access and Mobility Management Function,AMF)、会话管理功能(Session Management Function,SMF)、用户平面功能(User Plane Function,UPF)、策略控制功能(Policy Control Function,PCF)、策略与计费规则功能单元(Policy and Charging Rules Function,PCRF)、边缘应用服务发现功能(Edge Application Server Discovery Function,EASDF)、统一数据管理(Unified Data Management,UDM),统一数据仓储(Unified Data Repository,UDR)、归属用户服务器(Home Subscriber Server,HSS)、集中式网络配置(Centralized network configuration,CNC)、网络存储功能(Network Repository Function,NRF),网络开放功能(Network Exposure Function,NEF)、本地NEF(Local NEF,或L-NEF)、绑定支持功能(Binding Support Function,BSF)、应用功能(Application Function,AF)等。需要说明的是,在本申请实施例中仅以NR系统中的核心网设备为例进行介绍,并不限定核心网设备的具体类型。
未来的移动通信系统不仅可以向内为通信系统提供AI服务。也可以向外部应用提供AI服务。例如,为无人驾驶、无人机、机器人等通信网络外部应用提供AI服务。这些千行百业的外部应用有不同的服务质量等级。如果还基于相关通信系统中的服务质量指标作为外部服务的服务指标。则无法匹配外部应用的服务质量要求。因此,本申请实施例中提供了一种AI服务的处理方法,在发送AI服务请求时携带用于指示AI
服务质量要求的指示信息,使得移动通信系统可以接收AI服务质量要求,使得移动通信系统基于AI服务质量要求可以为外部应用提供质量更高的服务。
下面结合附图,通过一些实施例及其应用场景对本申请实施例提供的AI服务的处理方法进行详细地说明。
图2是本申请实施例提供的AI服务的处理方法的流程示意图之一。如图2所示,本实施例提供的方法,包括:
步骤101、网络侧设备接收来自目标设备的第一信息,第一信息包括第一服务的AI服务标识和用于指示AI服务质量要求的指示信息;AI服务质量要求包括以下至少一种类型的要求:安全性要求、舒适性要求、高效性要求和时延要求;安全性要求用于表征所述目标设备的防碰撞能力;舒适性要求用于表征目标设备的乘坐舒适度;高效性要求用于表征目标设备完成任务的效率;时延要求用于表征目标设备发起AI服务请求与获取到服务结果之间的时间差;
具体地,网络侧设备例如基站功能或核心网功能,目标设备例如终端,车辆。
目标设备向网络侧设备发送第一信息,例如第一信息可以是AI服务请求,网络侧设备接收该第一信息,第一信息包括:AI服务标识和用于指示AI服务质量要求的指示信息。
例如,目标设备请求的服务为环境感知,例如对车道线、静态物体和动态物体等的感知,或者目标设备请求的服务为路径规划,例如提供一条轨迹,包含多个轨迹点标识,每个轨迹点标识对应一个距离和速度;目标设备例如为车载终端,安全性要求可以是通过车辆与障碍物的距离计算的代价表示,舒适性要求可以是通过车辆行驶的路径朝向偏差、加速度等参数计算的代价表示,高效性要求用于表征目标设备完成任务的效率,例如通过路径行驶时长、与理想无障碍路径偏差、是否超过道路边界等参数计算的代价表示,时延要求用于表征目标设备发起AI服务请求与获取到网络侧设备反馈的服务结果之间的时间差,例如时延要求为小于100ms。
步骤102、网络侧设备基于AI服务标识和AI服务质量要求,向目标设备反馈第一服务的服务结果。
具体地,网络侧设备根据AI服务标识和AI服务质量要求,提供相应的服务,即获取该AI服务标识对应的第一服务的服务结果,例如基于AI模型获取服务结果,例如可以根据AI服务质量要求对AI模型的输出进行迭代,获取服务质量较好的服务结果,并把服务结果反馈给目标设备。
本实施例的方法,网络侧设备接收来自目标设备的第一信息,第一信息包括第一服务的AI服务标识和用于指示AI服务质量要求的指示信息;AI服务质量要求包括以下至少一种类型的要求:安全性要求、舒适性要求、高效性要求和时延要求;安全性要求用于表征所述目标设备的防碰撞能力;舒适性要求用于表征目标设备的乘坐舒适度;高效性要求用于表征目标设备完成任务的效率;时延要求用于表征目标设备发起AI服务请求与获取到服务结果之间的时间差;进一步地,网络侧设备基于AI服务标识和AI服务质量要求,获取第一服务的服务结果,并向目标设备反馈服务结果,即网络侧设备可以基于AI服务质量要求为目标设备提供服务,可以提升AI服务质量,从而实现了更好地在移动通信系统中支持AI服务。
可选地,网络侧设备获取用于确定质量评估方式的信息;用于确定质量评估方式的信息包括:质量评估方式对应的可执行文件和运行信息;质量评估方式用于对服务结果的质量进行评估。
可选地,运行信息包括以下至少一项:
运行环境配置信息、文件名称、输入格式、输出格式、指示预处理方式的信息、指示后处理方式的信息。
可选地,网络侧设备获取用于确定质量评估方式的信息,包括以下至少一项:
网络侧设备接收来自目标设备的用于确定质量评估方式的信息;
网络侧设备接收来自目标设备的第三方链接,并基于第三方链接获取用于确定质量评估方式的信息;
网络侧设备接收来自目标设备的应用程序编程接口(Application Programming Interface,API),并基于API获取用于确定质量评估方式的信息。
具体地,目标设备可以将用于确定质量评估方式的信息直接发送给网络侧设备,或通过第三方链接发送给网络侧设备,或通过API发送给网络侧设备,网络侧设备直接从接收到的信息中获取用于确定质量评估方式的信息,或,基于第三方链接或API获取用于确定质量评估方式的信息。
可选地,网络侧设备可以基于用于确定质量评估方式的信息,对服务结果进行评价,获取评价结果。
例如,还可以基于评价结果以及服务质量要求对提供的服务进行优化,例如优化AI模型。
可选地,在AI服务质量要求包括安全性要求的情况下,第一信息包括安全性要求指标门限和用于评估安全性质量的第一参数信息;其中,第一参数信息包括以下至少一项:
第一指示信息、第二信息、第一加权信息;
其中,第一指示信息用于指示安全性质量评估包括的至少一个第一目标参数项;第一加权信息包括各个所述第一目标参数项的加权值;第二信息为各个所述第一目标参数项与安全代价的函数映射信息,或者第一距离与安全等级的对应关系;
所述第一目标参数项包括以下至少一项:
第一距离、第一保护间隔门限、第二保护间隔门限;
其中,第一距离是目标设备与障碍物的距离;第一保护间隔门限用于表示最高安全等级的保护间隔下限,第二保护间隔门限用于表示最低安全等级的保护间隔下限。
具体地,函数映射信息例如用于指示与第一目标参数项对应的安全代价函数的标识,或安全代价函数的内容。
当第一距离大于保护间隔第一门限时,安全等级属于最高等级。当第一距离小于保护间隔第二门限时,会发生安全事故,不满足服务质量要求。
例如,评估安全性质量,即计算安全代价可以通过第一距离、第一保护间隔门限、第二保护间隔门限、第一距离与安全等级的对应关系等实现。
可选地,在第一距离位于第一保护间隔门限和第二保护间隔门限之间时,第一距离越大,安全等级越高。
例如,安全等级的数值越大,表示安全性越高;或,安全等级的数值越小,表示安全性越高。
在安全等级的数值越小,表示安全性越高的情况下,当第一距离位于第一门限和第二门限之间时,安全等级是第一距离的单调递减函数。
可选地,网络侧设备根据以下至少一项信息,确定第一距离;
至少一项信息包括:目标设备的尺寸、航向角、目标设备的位置和障碍物边界。
具体地,目标设备例如为车辆,航向角为该车辆的航向角。
示例性地,以无人车为例,第一距离为无人车边界到障碍物边界之间的距离。例如,无人车边界以覆盖无人车的三维度长方体来表示。同理,障碍物的边界也是由覆盖障碍物的三维长方体来表示。
第一保护间隔门限为10米,表示当第一距离大于或等于10米时,安全代价为0;
第二保护间隔门限为0.1米,表示当第一距离小于0.1米时,规划的轨迹无效。安全代价为负无穷大。
安全代价的函数标识为4,当第一距离大于或等于0.1米,且小于10米时,选择函数exp(-d/10),d为第一距离,来计算安全代价,如下表1所示:
表1
可选地,在AI服务质量要求包括舒适性要求的情况下,第一信息包括舒适性要求指标门限和用于评估舒适性质量的第二参数信息;其中,第二参数信息包括以下至少一项:
第二指示信息、第三信息、第二加权信息;
其中,第二指示信息用于指示舒适性质量评估包括的至少一个第二目标参数项,第三信息为各个所述第二目标参数项对应的舒适性代价函数的信息;第二加权信息包括各个所述第二目标参数项的加权值;所述第二目标参数项包括以下至少一项:
路径的朝向偏差、路径的航向角、路径的曲率、路径的曲率导数、路径的曲率变化率、加速度、加加速度。
具体地,第二指示信息用于指示用于舒适性质量评估的包括的目标参数项,例如,舒适性质量评估时考虑以下参数:路径的朝向偏差、航向角、曲率。
示例性地,第一信息中包括朝向偏差代价函数的信息(例如朝向偏差代价函数标识)、路径的曲率代价函数的信息(函数标识)和曲率变化率代价函数的信息(函数标识)。而这三者的代价函数是提前约定好的。三者的加权值也是提前约定的。如果三者的加权值与约定值不同,也可以携带在第一信息中,即通过第一信息指示。
例如,舒适性评估考虑参数:路径的朝向偏差、路径的曲率、曲率变化率;
假设总规划时长是15秒,时间间隔是1秒,则这个路径包括规划的15个轨迹点。
路径的朝向偏差为路径的一阶导数,朝向偏差代价的计算方法为,在每个时间间隔内,对路径的一阶导数求平方;再把15个时间间隔的值累加起来。计算方法可以是预定义的,也可以通过包含在第二信息的朝向偏差代价函数信息中。
路径的曲率为路径的二阶导数,路径的曲率代价的计算方法为,在每个时间间隔内,对路径的二阶导数求平方;再把15个时间间隔的值累加起来。
曲率变化率为路径的三阶导数,曲率变化率代价的计算方法为,在每个时间间隔内,对路径的三阶导数求平方;再把15个时间间隔的值累加起来。
朝向偏差代价、路径的曲率代价和曲率变化率代价的加权值分别为w1,w2和w3。
则舒适度总代价为:
w1×朝向偏差代价+w2×路径的曲率代价+w3×曲率变化率代价。
可选地,在AI服务质量要求包括所述高效性要求的情况下,第一信息求包括所述高效性要求指标门限和用于评估高效性质量的第三参数信息;其中,第三参数信息包括以下至少一项:
第三指示信息、第四信息、第三加权值;
其中,第三指示信息用于指示高效性质量评估包括的至少一个第三目标参数项,第四信息为各个所述第三目标参数项对应的高效性代价函数的信息;所述第三加权信息包括各个所述第三目标参数项的加权值;所述第三目标参数项包括以下至少一项:
与车道中心线的偏差、与理想无障碍路径的偏差、超过道路边界的惩罚值、路径行驶时长、路径行驶距离。
具体地,第三指示信息用于指示用于高效性质量评估的包括的目标参数项,例如,高效性质量评估时考虑以下参数:与车道中心线的偏差、路径行驶时长、路径行驶距离、超过道路边界的惩罚值。示例性地,第一信息中包括以下几项:
与车道中心线的偏差代价函数信息;(表示与理想无障碍路径偏差的惩罚函数是提前约定好的)
超过道路边界的惩罚值为100;(表示与预先约定的值不同,或者预先没有约定该值)
路径行驶时长代价函数信息;(表示已经提前约定了计算方法)
与车道中心线的偏差代价、超过道路边界的惩罚值和路径行驶时长代价的加权值分别为w1,w2和w3。
假设,总规划时长是15秒,时间间隔是1秒,则这个路径包括规划的15个轨迹点。
与理想无障碍路径偏差代价的计算方法为,在每个时间间隔内,对规划路径与理想无障碍路径的差值求平方;再把15个时间间隔的值累加起来。计算方法可以是预定义的,也可以通过包含在第三信息的朝向偏差代价函数信息中。
路径行驶时长代价的计算方法为1/(1+第二距离),其中第二距离为规划路径的总行驶距离。计算方法可以是预定义的,也可以通过包含在第三信息的朝向偏差代价函数信息中。
则高效性要求的总代价为:
w1×与理想无障碍路径偏差代价+w2×100+w3×1/(1+第二距离)。
可选地,AI服务质量要求还包括:
用于指示服务质量要求类型的指示信息,以及每个类型的服务质量要求的合并方式,所述合并方式包括以下至少一项:
每个类型的服务质量要求包括的参数项合并的加权函数信息;
每个类型的服务质量要求包括的参数项的加权值。
具体地,加权函数信息例如包括:加权函数的标识或加权函数的内容,例如安全性要求、舒适性要求、高效性要求和时延要求分别对应的加权函数,每一种类型中多个参数对应的加权值,可以获取到每个类型的服务质量要求。
可选地,综合服务质量要求为由第一信息确定的,包括以下至少一种方式:
基于第一信息直接得到的;
基于第一信息得到AI服务质量要求,再基于AI服务质量要求得到的。
具体地,综合服务质量要求可以是由第一信息直接得到,例如第一信息中直接包
括综合服务质量要求的内容,或包括计算综合服务质量要求的内容;或,先由第一信息得到AI服务质量要求,再基于AI服务质量要求得到的,例如第一信息包括多项服务质量要求,由该多项服务质量要求得到综合服务质量要求。
可选地,在综合服务质量要求为由第一信息确定的情况下,第一信息还包括以下至少一项:
综合服务质量等级;
综合服务质量门限;
各服务质量的权重;
各服务质量的加权方法。
可选地,综合服务质量等级与各服务质量等级对应关系,例如表2所示:
表2
可选地,综合服务质量等级与各服务质量数值(服务质量数值通过代价或时延表示)的对应关系,例如表3所示:
表3
示例性地,综合服务质量值可以由各服务质量数值映射得到。例如,涉及的质量要求指标包括安全性,高效性,舒适性和时延。
可选地,综合质量值计算方式如下:
综合服务质量值=a1×安全性服务质量数值+a2×高效性服务质量数值+a3×舒适性服务质量数值数据+a4×时延。其中,a1,a2,a3和a4分别为四者加权值。
可选地,综合服务质量等级与舒适性代价的对应关系,例如表4所示:
表4
可选地,综合服务质量等级与安全性代价的对应关系,例如表5所示:
表5
可选地,综合服务质量等级与高效性代价的对应关系,例如表6所示:
表6
可选地,综合服务质量等级与时延的对应关系,例如表7所示:
表7
可选地,AI服务质量要求通过以下至少一项表示:
质量等级、质量要求指标门限;质量要求指标门限包括以下至少一项:安全性要求指标门限、舒适性要求指标门限、高效性要求指标门限和时延要求指标门限。
可选地,AI服务质量要求还包括以下至少一种类型的要求:
速度保持代价;
路径单调性;
目标设备速度上限;
路径规划的总时长;
路径规划的时间间隔;
其中,速度保持代价为基于服务结果中包括的行驶速度与目标行驶速度的差值得到的,目标行驶速度为由第二信息确定的;第二信息包括以下至少一项:路径的曲率、交通规则;
路径单调性是指在一次规划的路径中目标设备与路径起点的距离是单调递增的,或目标设备与路径终点的距离是单调递减的;
目标设备速度上限为基于交通规则和/或目标设备动力学得到的。
具体地,速度保持代价可以通过网络侧设备计算得到的行驶速度,与目标行驶速度的差值得到的,偏离目标行驶速度会受到惩罚。目标行驶速度由路径的曲率、交通规则等确定,例如基于道路速度限制、曲率和其他交通规则等来确定。上述速度保持代价、路径单调性、目标设备速度上限等可用于网络侧设备为目标设备提供服务,例如上述信息作为网络侧设备的AI模型的输入数据,可以提高服务质量。
可选地,第一信息中还包括:用于计算速度保持代价的信息,用于计算速度保持代价的信息包括以下至少一项:
偏离速度惩罚函数指示信息;
偏离速度的惩罚倍数。
可选地,偏离速度惩罚函数指示信息,例如为偏离速度惩罚函数的具体内容,或函数标识。惩罚倍数可以为计算速度保持代价的权值。例如,偏离速度惩罚函数为f,则速度保持代价为f(偏离速度),或者惩罚倍数为αp,则速度保持代价为αp×偏离速度,其中,f例如为服务结果中包括的行驶速度与目标行驶速度的差值的函数。
可选地,第一信息中还包括以下至少一项:
目标设备的尺寸;
目标设备的位置;
目标设备对周围物体的感知结果;感知结果包括以下至少一项的感知结果:类型、距离、速度、加速度、边界;
目标设备对周围物体的原始感知数据。
具体地,目标设备例如为车辆,通过第一信息中包括的目标设备的尺寸、目标设备的位置、原始感知数据、感知结果等,可以辅助网络侧设备进行AI模型的推理或计算,以提升网络侧设备提供的服务质量。
可选地,该方法还包括:
网络侧设备基于第一信息和质量评估方式的信息,或,基于第一信息确定第六信息,第六信息包括以下至少一项:质量要求指标、质量要求指标计算方法、质量要求指标输入数据和格式、质量要求指标门限。
具体地,质量要求指标例如包括以下至少一项:安全性、舒适性、高效性和时延指标,基于第六信息可以获取服务结果,为目标设备提供服务。或者,基于第六信息,还可以获取评价结果。
可选地,该方法还包括:
网络侧设备基于第六信息和服务结果,确定服务结果的第一评价结果;
网络侧设备向所述目标设备发送服务结果的第一评价结果。
具体地,网络侧设备获取到服务结果之后,可以发送给目标设备,可选地,还可以将服务结果的第一评价结果发送给目标设备,例如目标设备可以基于该第一评价结果对服务结果进行筛选,选择满足要求的服务结果,基于服务结果执行相应的任务。
可选地,目标设备还可以向网络侧设备发送服务结果的第二评价结果,网络侧设备接收目标设备发送的第二评价结果,第二评价结果为所述目标设备对服务结果进行评价的评价结果。
第二评价结果可以用于网络侧设备优化提供的服务,例如,网络侧设备可以基于目标设备反馈的第二评价结果对AI模型进行优化。
可选地,第一评价结果或第二评价结果包括以下至少一项:
AI服务标识;
AI服务标识对应的服务等级;
AI服务标识对应的服务质量数值;
综合服务质量数值;
综合服务质量等级。
具体地,例如,AI服务标识表示当前服务为环境感知,当前评价结果例如为该环境感知的服务结果的服务等级为5级,例如服务质量要求中服务等级为4级,数值越小,等级越高,说明提供的服务结果不满足服务质量要求,需要重新进行优化。其它参数类似。
可选地,步骤102可以通过如下方式实现:
网络侧设备基于AI服务标识和第七信息,确定提供服务的算法信息;第七信息包括以下至少一项:AI服务质量要求、综合服务质量要求;
网络侧设备基于AI服务标识以及算法信息,向目标设备反馈服务结果。
具体地,网络侧设备基于AI服务标识和第七信息,确定提供服务的算法信息,基于算法信息执行服务算法(例如AI模型推理),获取服务结果,并向目标设备反馈服
务结果。
可选地,在提供的AI服务为训练服务的情况下,算法信息包括训练的奖励函数;
在提供的AI服务为推理服务的情况下,算法信息包括推理模型。
可选地,该方法还包括:
网络侧设备基于AI服务标识、推理服务对应的服务结果和综合服务质量要求,确定推理服务对应的服务结果的第一评价结果。
例如推理服务对应的服务结果包括:目标设备预设时间范围内的速度、路径等,综合服务质量要求包括综合服务质量等级的要求,利用该速度和目标行驶速度计算速度保持代价,基于速度保持代价可以确定该服务结果的综合服务质量等级,例如速度保持代价越小,综合服务质量等级越高,假设速度保持代价对应的综合服务质量等级为3级,但是综合服务质量等级的要求为5级,此时假设安全等级的数值越大,等级越高,则说明不满足综合服务质量要求。
示例性地,目标设备为车辆,网络侧设备为基站,当前服务为轨迹规划服务。
服务质量要求已经预先定义好,无需额外指示。可以根据服务标识直接获得。
如图3所示,该方法包括如下步骤:
步骤1:车辆向基站发送服务标识,该服务标识用于指示路径规划服务。
步骤2:基站根据服务标识,获取第六信息。第六信息包括以下至少一项:质量要求指标,质量要求指标计算方法,质量要求指标输入数据和格式,质量要求指标门限。
例如,确定质量要求指标包括安全性、舒适性、高效性和时延。
质量要求指标计算方法,质量要求指标输入数据和格式已经预先定义好。
质量要求指标门限为综合服务质量门限,例如门限为10。当门限小于10时,则满足服务质量要求。
步骤3:基于服务标识,确定服务算法。执行服务算法(例如AI模型推理),获取服务结果,再基于质量要求指标计算方法验证服务结果是否满足服务要求。或选择服务质量最高的一个服务结果。
步骤4:反馈服务结果和评价结果。评价结果可以是一个综合服务质量值,或者一个等级。
可选地,网络侧设备根据第一信息,确定综合服务质量要求,进而确定服务算法。
此时,质量要求指标,质量要求指标计算方法,质量要求指标输入数据和格式都是预先定义的。
例如,质量要求指标包括:舒适性、高效性、安全性和时延
质量要求指标计算方法,质量要求指标输入数据和格式都是预先定义的。如表8所示:
表8
上述实施方式中,通过提供AI服务质量要求,使得网络侧设备可以获得更完整更丰富的信息,从而提升服务质量,目标设备通过享受网络侧设备提供的服务,降低了自身设备的复杂度,进一步,网络侧设备由于可以识别不同的AI服务质量要求,从而可以为目标设备提供更灵活的服务。
图4是本申请实施例提供的AI服务的处理方法的流程示意图之二。如图3、图4所示,本实施例提供的方法,包括:
步骤201、目标设备向网络侧设备发送第一信息,第一信息包括第一服务的AI服务标识和用于指示AI服务质量要求的指示信息;AI服务质量要求包括以下至少一种类型的要求:安全性要求、舒适性要求、高效性要求和时延要求;
其中,安全性要求用于表征目标设备的防碰撞能力;所述舒适性要求用于表征所述目标设备的乘坐舒适度;所述高效性要求用于表征所述目标设备完成任务的效率;所述时延要求用于表征所述目标设备发起AI服务请求与获取到服务结果之间的时间差;
步骤202、目标设备接收网络侧设备发送的第一服务的服务结果。
可选地,所述指示信息通过以下至少一项表示所述AI服务质量要求:
质量等级、质量要求指标门限;其中,所述质量要求指标门限包括以下至少一项:安全性要求指标门限、舒适性要求指标门限、高效性要求指标门限和时延要求指标门限。
可选地,所述方法还包括:
所述目标设备向所述网络侧设备发送用于确定质量评估方式的信息;所述质量评估方式用于对服务结果的质量进行评估;所述用于确定质量评估方式的信息包括:质量评估方式对应的可执行文件和运行信息。
可选地,所述运行信息包括以下至少一项:
运行环境配置信息、文件名称、输入格式、输出格式、指示预处理方式的信息、指示后处理方式的信息。
可选地,所述目标设备向所述网络侧设备发送质量评估方式的信息,包括以下至少一项:
所述目标设备向所述网络侧设备发送第三方链接,所述第三方链接用于获取所述用于确定质量评估方式的信息;
所述目标设备向所述网络侧设备发送应用程序编程接口API,所述应用程序编程接口API用于获取所述用于确定质量评估方式的信息。
可选地,在所述AI服务质量要求包括所述安全性要求的情况下,所述第一信息包括所述安全性要求指标门限和用于评估安全性质量的第一参数信息;其中,第一参数信息包括以下至少一项:
第一指示信息、第二信息、第一加权信息;
其中,第一指示信息用于指示舒适性质量评估包括的至少一个第一目标参数项;所述第一加权信息包括各个所述第一目标参数项的加权值;
所述第一目标参数项包括以下至少一项:
第一距离、第一保护间隔门限、第二保护间隔门限;
其中,第一距离是所述目标设备与障碍物的距离;所述第一保护间隔门限用于表
示最高安全等级的保护间隔下限,所述第二保护间隔门限用于表示最低安全等级的保护间隔下限;
第二信息为各个所述第一目标参数项与安全代价的函数映射信息,或者第一距离与安全等级的对应关系。
可选地,在所述第一距离位于第一保护间隔门限和第二保护间隔门限之间时,安全等级越高,所述第一距离越大。
可选地,所述第一距离为基于以下至少一项信息确定的;
所述至少一项信息包括:目标设备的尺寸、航向角、目标设备的位置和障碍物边界。
可选地,在所述AI服务质量要求包括所述舒适性要求的情况下,所述第一信息包括所述舒适性要求指标门限和用于评估舒适性质量的第二参数信息;其中,所述第二参数信息包括以下至少一项:
第二指示信息、第三信息、第二加权信息;
其中,第二指示信息用于指示舒适性质量评估包括的至少一个第二目标参数项,第三信息为各个所述第二目标参数项对应的舒适性代价函数的信息;所述第二加权信息包括各个所述第二目标参数项的加权值;所述第二目标参数项包括以下至少一项:
路径的朝向偏差、路径的航向角、路径的曲率、路径的曲率导数、路径的曲率变化率、加速度、加加速度。
可选地,在所述AI服务质量要求包括所述高效性要求的情况下,所述第一信息包括所述高效性要求指标门限和用于评估高效性质量的第三参数信息;其中,所述第三参数信息包括以下至少一项:
第三指示信息、第四信息、第三加权信息;
其中,第三指示信息用于指示高效性质量评估包括的至少一个第三目标参数项,第四信息为各个所述第三目标参数项对应的高效性代价函数的信息;所述第三加权信息包括各个所述第三目标参数项的加权值;所述第三目标参数项包括以下至少一项:
与车道中心线的偏差、与理想无障碍路径的偏差、超过道路边界的惩罚值、路径行驶时长、路径行驶距离。
可选地,所述AI服务质量要求还包括以下至少一种类型的要求:
速度保持代价;
路径单调性;
目标设备速度上限;
路径规划的总时长;
路径规划的时间间隔;
其中,速度保持代价为基于服务结果中包括的行驶速度与目标行驶速度的差值得到的,目标行驶速度为由第五信息确定的;所述第五信息包括以下至少一项:路径的曲率、交通规则;
路径单调性是指在一次规划的路径中所述目标设备与路径起点的距离是单调递增的,或所述目标设备与路径终点的距离是单调递减的;
目标设备速度上限为基于交通规则和/或目标设备动力学得到的。
可选地,所述第一信息中还包括:用于计算速度保持代价的信息,所述用于计算速度保持代价的信息包括以下至少一项:
偏离速度惩罚函数信息;
偏离速度的惩罚倍数。
可选地,所述AI服务质量要求还包括用于指示服务质量要求类型的指示信息,以及每个类型的服务质量要求的合并方式,所述合并方式包括以下至少一项:
每个类型的服务质量要求包括的参数项合并的加权函数信息;
每个类型的服务质量要求包括的参数项的加权值。
可选地,第一信息中还包括以下至少一项:
目标设备的尺寸;
目标设备的位置;
目标设备对周围物体的感知结果;所述感知结果包括以下至少一项的感知结果:类型、距离、速度、加速度、边界;
目标设备对周围物体的原始感知数据。
可选地,所述方法还包括:
所述目标设备接收所述网络侧设备发送的所述服务结果的第一评价结果;所述服务结果的第一评价结果为基于所述第六信息和所述服务结果确定的;所述第六信息包括以下至少一项:质量要求指标、质量要求指标计算方法、质量要求指标输入数据和格式、质量要求指标门限。
可选地,所述方法还包括:
所述目标设备向所述网络侧设备发送第二评价结果,所述第二评价结果为所述目标设备对所述服务结果进行评价的评价结果。
可选地,第一评价结果或第二评价结果包括以下至少一项:
AI服务标识;
AI服务标识对应的服务等级;
AI服务标识对应的服务质量数值;
综合服务质量数值;
综合服务质量等级。
可选地,综合服务质量要求为由所述第一信息确定的,包括以下至少一种方式:
基于所述第一信息直接得到的;
基于所述第一信息得到AI服务质量要求,再基于所述AI服务质量要求得到的。
可选地,在所述综合服务质量要求为由所述第一信息确定的情况下,所述第一信息还包括以下至少一项:
综合服务质量等级;
综合服务质量门限;
各服务质量的权重;
各服务质量的加权方法。
本实施例的方法,其具体实现过程与技术效果与网络侧方法实施例中类似,具体可以参见终端侧方法实施例中的详细介绍,此处不再赘述。
本申请实施例提供的AI服务的处理方法,执行主体可以为AI服务的处理装置。本申请实施例中以AI服务的处理装置执行AI服务的处理方法为例,说明本申请实施例提供的AI服务的处理装置。
图5是本申请提供的AI服务的处理装置的结构示意图之一。如图5所示,本实施例提供的AI服务的处理装置,包括:
接收模块210,用于接收来自目标设备的第一信息,所述第一信息包括第一服务
的AI服务标识和用于指示AI服务质量要求的指示信息;所述AI服务质量要求包括以下至少一种类型的要求:安全性要求、舒适性要求、高效性要求和时延要求;所述安全性要求用于表征所述目标设备的防碰撞能力;所述舒适性要求用于表征所述目标设备的乘坐舒适度;所述高效性要求用于表征所述目标设备完成任务的效率;所述时延要求用于表征所述目标设备发起AI服务请求与获取到服务结果之间的时间差;
处理模块220,用于基于所述AI服务标识和AI服务质量要求,向目标设备反馈第一服务的服务结果。
可选地,所述指示信息通过以下至少一项表示所述AI服务质量要求:
质量等级、质量要求指标门限;其中,所述质量要求指标门限包括以下至少一项:安全性要求指标门限、舒适性要求指标门限、高效性要求指标门限和时延要求指标门限。
可选地,接收模块210,还用于:
获取用于确定质量评估方式的信息;所述质量评估方式用于对服务结果的质量进行评估;所述用于确定质量评估方式的信息包括:质量评估方式对应的可执行文件和运行信息。
可选地,所述运行信息包括以下至少一项:
运行环境配置信息、文件名称、输入格式、输出格式、指示预处理方式的信息、指示后处理方式的信息。
可选地,所述接收模块210,具体用于执行以下至少一项:
接收来自所述目标设备的所述用于确定质量评估方式的信息;
接收来自所述目标设备的第三方链接,并基于所述第三方链接获取所述用于确定质量评估方式的信息;
接收来自所述目标设备的应用程序编程接口API,并基于所述API获取所述用于确定质量评估方式的信息。
可选地,在所述AI服务质量要求包括所述安全性要求的情况下,所述第一信息包括所述安全性要求指标门限和用于评估安全性质量的第一参数信息;其中,第一参数信息包括以下至少一项:
第一指示信息、第二信息、第一加权信息;
其中,第一指示信息用于指示舒适性质量评估包括的至少一个第一目标参数项;所述第一加权信息包括各个所述第一目标参数项的加权值;
所述第一目标参数项包括以下至少一项:
第一距离、第一保护间隔门限、第二保护间隔门限;
其中,第一距离是所述目标设备与障碍物的距离;所述第一保护间隔门限用于表示最高安全等级的保护间隔下限,所述第二保护间隔门限用于表示最低安全等级的保护间隔下限;
第二信息为各个所述第一目标参数项与安全代价的函数映射信息,或者第一距离与安全等级的对应关系。
可选地,在所述第一距离位于第一保护间隔门限和第二保护间隔门限之间时,安全等级越高,所述第一距离越大。
可选地,所述处理模块220还用于:
根据以下至少一项信息,确定所述第一距离;
所述至少一项信息包括:目标设备的尺寸、航向角、目标设备的位置和障碍物边
界。
可选地,在所述AI服务质量要求包括所述舒适性要求的情况下,所述第一信息包括所述舒适性要求指标门限和用于评估舒适性质量的第二参数信息;其中,所述第二参数信息包括以下至少一项:
第二指示信息、第三信息、第二加权信息;
其中,第二指示信息用于指示舒适性质量评估包括的至少一个第二目标参数项,第三信息为各个所述第二目标参数项对应的舒适性代价函数的信息;所述第二加权信息包括各个所述第二目标参数项的加权值;所述第二目标参数项包括以下至少一项:
路径的朝向偏差、路径的航向角、路径的曲率、路径的曲率导数、路径的曲率变化率、加速度、加加速度。
可选地,在所述AI服务质量要求包括所述高效性要求的情况下,所述第一信息包括所述高效性要求指标门限和用于高效性质量评估的第三参数信息;其中,所述第三参数信息包括以下至少一项:
第三指示信息、第四信息、第三加权信息;
其中,第三指示信息用于指示高效性质量评估包括的至少一个第三目标参数项,第四信息为各个所述第三目标参数项对应的高效性代价函数的信息;所述第三加权信息包括各个所述第三目标参数项的加权值;所述第三目标参数项包括以下至少一项:
与车道中心线的偏差、与理想无障碍路径的偏差、超过道路边界的惩罚值、路径行驶时长、路径行驶距离。
可选地,所述AI服务质量要求还包括以下至少一种类型的要求:
速度保持代价;
路径单调性;
目标设备速度上限;
路径规划的总时长;
路径规划的时间间隔;
其中,速度保持代价为基于服务结果中包括的行驶速度与目标行驶速度的差值得到的,目标行驶速度为由第五信息确定的;所述第五信息包括以下至少一项:路径的曲率、交通规则;
路径单调性是指在一次规划的路径中所述目标设备与路径起点的距离是单调递增的,或所述目标设备与路径终点的距离是单调递减的;
目标设备速度上限为基于交通规则和/或目标设备动力学得到的。
可选地,所述第一信息中还包括:用于计算速度保持代价的信息,所述用于计算速度保持代价的信息包括以下至少一项:
偏离速度惩罚函数信息;
偏离速度的惩罚倍数。
可选地,所述AI服务质量要求还包括:用于指示服务质量要求类型的指示信息,以及每个类型的服务质量要求的合并方式,所述合并方式包括以下至少一项:
每个类型的服务质量要求包括的参数项合并的加权函数信息;
每个类型的服务质量要求包括的参数项的加权值。
可选地,第一信息中还包括以下至少一项:
目标设备的尺寸;
目标设备的位置;
目标设备对周围物体的感知结果;所述感知结果包括以下至少一项的感知结果:类型、距离、速度、加速度、边界;
目标设备对周围物体的原始感知数据。
可选地,所述处理模块220还用于:
基于所述第一信息和质量评估方式的信息,或,基于所述第一信息确定第三信息,所述第三信息包括以下至少一项:质量要求指标、质量要求指标计算方法、质量要求指标输入数据和格式、质量要求指标门限。
可选地,所述处理模块220还用于:
基于所述第三信息和所述服务结果,确定所述服务结果的第一评价结果;
所述网络侧设备向所述目标设备发送所述服务结果的第一评价结果。
可选地,所述接收模块210还用于:
接收所述目标设备发送的第二评价结果,所述第二评价结果为所述目标设备对所述服务结果进行评价的评价结果。
可选地,第一评价结果或第二评价结果包括以下至少一项:
AI服务标识;
AI服务标识对应的服务等级;
AI服务标识对应的服务质量数值;
综合服务质量数值;
综合服务质量等级。
可选地,所述处理模块220具体用于:
基于所述AI服务标识和第四信息,确定提供服务的算法信息;所述第四信息包括以下至少一项:AI服务质量要求、综合服务质量要求;
基于所述AI服务标识以及所述算法信息,向目标设备反馈所述服务结果。
可选地,在提供的AI服务为训练服务的情况下,所述算法信息包括训练的奖励函数;
在提供的AI服务为推理服务的情况下,所述算法信息包括推理模型。
可选地,所述处理模块220还用于:
基于所述AI服务标识、推理服务对应的服务结果和综合服务质量要求,确定推理服务对应的服务结果的第一评价结果。
可选地,所述综合服务质量要求为由所述第一信息确定的,包括以下至少一种方式:
基于所述第一信息直接得到的;
基于所述第一信息得到AI服务质量要求,再基于所述AI服务质量要求得到的。
可选地,在所述综合服务质量要求为由所述第一信息确定的情况下,所述第一信息还包括以下至少一项:
综合服务质量等级;
综合服务质量门限;
各服务质量的权重;
各服务质量的加权方法。
本实施例的装置,可以用于执行前述网络侧方法实施例中任一实施例的方法,其具体实现过程与技术效果与网络侧方法实施例中类似,具体可以参见网络侧方法实施例中的详细介绍,此处不再赘述。
图6是本申请提供的AI服务的处理装置的结构示意图之二。如图6所示,本实施例提供的AI服务的处理装置,包括:
发送模块310,用于向网络侧设备发送第一信息,所述第一信息包括第一服务的AI服务标识和用于指示AI服务质量要求的指示信息;所述AI服务质量要求包括以下至少一种类型的要求:安全性要求、舒适性要求、高效性要求和时延要求;所述安全性要求用于表征所述目标设备的防碰撞能力;所述舒适性要求用于表征所述目标设备的乘坐舒适度;所述高效性要求用于表征所述目标设备完成任务的效率;所述时延要求用于表征所述目标设备发起AI服务请求与获取到服务结果之间的时间差;
接收模块320,用于接收所述网络侧设备发送的第一服务的服务结果。
可选地,所述指示信息通过以下至少一项表示所述AI服务质量要求:
质量等级、质量要求指标门限;其中,所述质量要求指标门限包括以下至少一项:安全性要求指标门限、舒适性要求指标门限、高效性要求指标门限和时延要求指标门限。
可选地,所述发送模块310,还用于:
向所述网络侧设备发送用于确定质量评估方式的信息;所述质量评估方式用于对服务结果的质量进行评估;所述用于确定质量评估方式的信息包括:质量评估方式对应的可执行文件和运行信息。
可选地,所述运行信息包括以下至少一项:
运行环境配置信息、文件名称、输入格式、输出格式、指示预处理方式的信息、指示后处理方式的信息。
可选地,所述发送模块310,具体用于执行以下至少一项:
向所述网络侧设备发送第三方链接,所述第三方链接用于获取所述用于确定质量评估方式的信息;
向所述网络侧设备发送应用程序编程接口API,所述应用程序编程接口API用于获取所述用于确定质量评估方式的信息。
可选地,在所述AI服务质量要求包括所述安全性要求的情况下,所述第一信息包括所述安全性要求指标门限和用于评估安全性质量的第一参数信息;其中,第一参数信息包括以下至少一项:
第一指示信息、第二信息、第一加权信息;
其中,第一指示信息用于指示舒适性质量评估包括的至少一个第一目标参数项;所述第一加权信息包括各个所述第一目标参数项的加权值;
所述第一目标参数项包括以下至少一项:
第一距离、第一保护间隔门限、第二保护间隔门限;
其中,第一距离是所述目标设备与障碍物的距离;所述第一保护间隔门限用于表示最高安全等级的保护间隔下限,所述第二保护间隔门限用于表示最低安全等级的保护间隔下限;
第二信息为各个所述第一目标参数项与安全代价的函数映射信息,或者第一距离与安全等级的对应关系。
可选地,在所述第一距离位于第一保护间隔门限和第二保护间隔门限之间时,安全等级越高,所述第一距离越大。
可选地,所述第一距离为基于以下至少一项信息确定的;
所述至少一项信息包括:目标设备的尺寸、航向角、目标设备的位置和障碍物边
界。
可选地,在所述AI服务质量要求包括所述舒适性要求的情况下,所述第一信息包括所述舒适性要求指标门限和用于评估舒适性质量的第二参数信息;其中,所述第二参数信息包括以下至少一项:
第二指示信息、第三信息、第二加权信息;
其中,第二指示信息用于指示舒适性质量评估包括的至少一个第二目标参数项,第三信息为各个所述第二目标参数项对应的舒适性代价函数的信息;所述第二加权信息包括各个所述第二目标参数项的加权值;所述第二目标参数项包括以下至少一项:
路径的朝向偏差、路径的航向角、路径的曲率、路径的曲率导数、路径的曲率变化率、加速度、加加速度。
可选地,在所述AI服务质量要求包括所述高效性要求的情况下,所述第一信息包括所述高效性要求指标门限和用于高效性质量评估的第三参数信息;其中,所述第三参数信息包括以下至少一项:
第三指示信息、第四信息、第三加权信息;
其中,第三指示信息用于指示高效性质量评估包括的至少一个第三目标参数项,第四信息为各个所述第三目标参数项对应的高效性代价函数的信息;所述第三加权信息包括各个所述第三目标参数项的加权值;所述第三目标参数项包括以下至少一项:
与车道中心线的偏差、与理想无障碍路径的偏差、超过道路边界的惩罚值、路径行驶时长、路径行驶距离。
可选地,所述AI服务质量要求还包括以下至少一种类型的要求:
速度保持代价;
路径单调性;
目标设备速度上限;
路径规划的总时长;
路径规划的时间间隔;
其中,速度保持代价为基于服务结果中包括的行驶速度与目标行驶速度的差值得到的,目标行驶速度为由第五信息确定的;所述第五信息包括以下至少一项:路径的曲率、交通规则;
路径单调性是指在一次规划的路径中所述目标设备与路径起点的距离是单调递增的,或所述目标设备与路径终点的距离是单调递减的;
目标设备速度上限为基于交通规则和/或目标设备动力学得到的。
可选地,所述第一信息中还包括:用于计算速度保持代价的信息,所述用于计算速度保持代价的信息包括以下至少一项:
偏离速度惩罚函数信息;
偏离速度的惩罚倍数。
可选地,所述AI服务质量要求还包括用于指示服务质量要求类型的指示信息,以及每个类型的服务质量要求的合并方式,所述合并方式包括以下至少一项:
每个类型的服务质量要求包括的参数项合并的加权函数信息;
每个类型的服务质量要求包括的参数项的加权值。
可选地,第一信息中还包括以下至少一项:
目标设备的尺寸;
目标设备的位置;
目标设备对周围物体的感知结果;所述感知结果包括以下至少一项的感知结果:类型、距离、速度、加速度、边界;
目标设备对周围物体的原始感知数据。
可选地,所述接收模块320还用于:
接收所述网络侧设备发送的所述服务结果的第一评价结果;所述服务结果的第一评价结果为基于所述第六信息和所述服务结果确定的;所述第六信息包括以下至少一项:质量要求指标、质量要求指标计算方法、质量要求指标输入数据和格式、质量要求指标门限。
可选地,所述发送模块310还用于:
向所述网络侧设备发送第二评价结果,所述第二评价结果为所述目标设备对所述服务结果进行评价的评价结果。
可选地,第一评价结果或第二评价结果包括以下至少一项:
AI服务标识;
AI服务标识对应的服务等级;
AI服务标识对应的服务质量数值;
综合服务质量数值;
综合服务质量等级。
可选地,综合服务质量要求为由所述第一信息确定的,包括以下至少一种方式:
基于所述第一信息直接得到的;
基于所述第一信息得到AI服务质量要求,再基于所述AI服务质量要求得到的。
可选地,在所述综合服务质量要求为由所述第一信息确定的情况下,所述第一信息还包括以下至少一项:
综合服务质量等级;
综合服务质量门限;
各服务质量的权重;
各服务质量的加权方法。
本实施例的装置,可以用于执行前述目标设备侧方法实施例中任一实施例的方法,其具体实现过程与技术效果与目标设备侧方法实施例中类似,具体可以参见目标设备侧方法实施例中的详细介绍,此处不再赘述。
本申请实施例中的AI服务的处理装置可以是电子设备,例如具有操作系统的电子设备,也可以是电子设备中的部件,例如集成电路或芯片。该电子设备可以是终端,也可以为除终端之外的其他设备。示例性的,终端可以包括但不限于上述所列举的终端11的类型,其他设备可以为服务器、网络附属存储器(Network Attached Storage,NAS)等,本申请实施例不作具体限定。
本申请实施例提供的AI服务的处理装置能够实现图2至图4的方法实施例实现的各个过程,并达到相同的技术效果,为避免重复,这里不再赘述。
可选地,如图7所示,本申请实施例还提供一种通信设备700,包括处理器701和存储器702,存储器702上存储有可在所述处理器701上运行的程序或指令,例如,该通信设备700为目标设备时,该程序或指令被处理器701执行时实现上述AI服务的处理方法实施例的各个步骤,且能达到相同的技术效果。该通信设备700为网络侧设备时,该程序或指令被处理器701执行时实现上述AI服务的处理方法实施例的各个步骤,且能达到相同的技术效果,为避免重复,这里不再赘述。
本申请实施例还提供一种目标设备,包括处理器和通信接口,所述通信接口用于向网络侧设备发送第一信息,所述第一信息包括第一服务的AI服务标识和用于指示AI服务质量要求的指示信息;所述AI服务质量要求包括以下至少一种类型的要求:安全性要求、舒适性要求、高效性要求和时延要求;所述安全性要求用于表征所述目标设备的防碰撞能力;所述舒适性要求用于表征所述目标设备的乘坐舒适度;所述高效性要求用于表征所述目标设备完成任务的效率;所述时延要求用于表征所述目标设备发起AI服务请求与获取到服务结果之间的时间差;所述通信接口还用于接收所述网络侧设备发送的第一服务的服务结果。该目标设备实施例与上述目标设备侧方法实施例对应,上述方法实施例的各个实施过程和实现方式均可适用于该终端实施例中,且能达到相同的技术效果。目标设备具体地,图8为实现本申请实施例的一种终端的硬件结构示意图。
该终端1000包括但不限于:射频单元1001、网络模块1002、音频输出单元1003、输入单元1004、传感器1005、显示单元1006、用户输入单元1007、接口单元1008、存储器1009、以及处理器1010等中的至少部分部件。
本领域技术人员可以理解,终端1000还可以包括给各个部件供电的电源(比如电池),电源可以通过电源管理系统与处理器1010逻辑相连,从而通过电源管理系统实现管理充电、放电、以及功耗管理等功能。图8中示出的终端结构并不构成对终端的限定,终端可以包括比图示更多或更少的部件,或者组合某些部件,或者不同的部件布置,在此不再赘述。
应理解的是,本申请实施例中,输入单元1004可以包括图形处理单元(Graphics Processing Unit,GPU)10041和麦克风10042,图形处理器10041对在视频捕获模式或图像捕获模式中由图像捕获装置(如摄像头)获得的静态图片或视频的图像数据进行处理。显示单元1006可包括显示面板10061,可以采用液晶显示器、有机发光二极管等形式来配置显示面板10061。用户输入单元1007包括触控面板10071以及其它输入设备10072中的至少一种。触控面板10071,也称为触摸屏。触控面板10071可包括触摸检测装置和触摸控制器两个部分。其它输入设备10072可以包括但不限于物理键盘、功能键(比如音量控制按键、开关按键等)、轨迹球、鼠标、操作杆,在此不再赘述。
本申请实施例中,射频单元1001将接收来自网络侧设备的下行数据接收后,可以传输给处理器1010进行处理;另外,射频单元1001可以将上行的数据发送给向网络侧设备发送上行数据。通常,射频单元1001包括但不限于天线、至少一个放大器、收发信机、耦合器、低噪声放大器、双工器等。
存储器1009可用于存储软件程序或指令以及各种数据。存储器1009可主要包括存储程序或指令的第一存储区和存储数据的第二存储区,其中,第一存储程序或指令区可存储操作系统、至少一个功能所需的应用程序或指令(比如声音播放功能、图像播放功能等)等。此外,存储器1009可以包括易失性存储器或非易失性存储器,或者,存储器1009可以包括易失性和非易失性存储器两者。包括高速随机存取存储器,还可以包括非易失性存储器,其中,非易失性存储器可以是只读存储器(Read-Only Memory,ROM)、可编程只读存储器(Programmable ROM,PROM)、可擦除可编程只读存储器(Erasable PROM,EPROM)、电可擦除可编程只读存储器(Electrically EPROM,EEPROM)或闪存。易失性存储器可以是随机存取存储器(Random Access Memory,RAM),静态随机存取存储器(Static RAM,SRAM)、动态随机存取存储器(Dynamic
RAM,DRAM)、同步动态随机存取存储器(Synchronous DRAM,SDRAM)、双倍数据速率同步动态随机存取存储器(Double Data Rate SDRAM,DDRSDRAM)、增强型同步动态随机存取存储器(Enhanced SDRAM,ESDRAM)、同步连接动态随机存取存储器(Synch link DRAM,SLDRAM)和直接内存总线随机存取存储器(Direct Rambus RAM,DRRAM)。本申请实施例中的存储器1009包括但不限于这些和任意其它适合类型的存储器例如至少一个磁盘存储器件、闪存器件、或其它非易失性固态存储器件。
处理器1010可包括一个或多个处理单元;可选的,处理器1010可集成应用处理器和调制解调处理器,其中,应用处理器主要处理涉及操作系统、用户界面和应用程序或指令等的操作,调制解调处理器主要处理无线通信信号,如基带处理器。可以理解的是,上述调制解调处理器也可以不集成到处理器1010中。
其中,射频单元1001,用于向网络侧设备发送第一信息,所述第一信息包括第一服务的AI服务标识和用于指示AI服务质量要求的指示信息;所述AI服务质量要求包括以下至少一种类型的要求:安全性要求、舒适性要求、高效性要求和时延要求;所述安全性要求用于表征所述目标设备的防碰撞能力;所述舒适性要求用于表征所述目标设备的乘坐舒适度;所述高效性要求用于表征所述目标设备完成任务的效率;所述时延要求用于表征所述目标设备发起AI服务请求与获取到服务结果之间的时间差;
射频单元1001,还用于接收所述网络侧设备发送的第一服务的服务结果。
可选地,所述指示信息通过以下至少一项表示所述AI服务质量要求:
质量等级、质量要求指标门限;其中,所述质量要求指标门限包括以下至少一项:安全性要求指标门限、舒适性要求指标门限、高效性要求指标门限和时延要求指标门限。
可选地,所述射频单元1001,还用于:
向所述网络侧设备发送用于确定质量评估方式的信息;所述质量评估方式用于对服务结果的质量进行评估;所述用于确定质量评估方式的信息包括:质量评估方式对应的可执行文件和运行信息。
可选地,所述运行信息包括以下至少一项:
运行环境配置信息、文件名称、输入格式、输出格式、指示预处理方式的信息、指示后处理方式的信息。
可选地,所述射频单元1001,具体用于执行以下至少一项:
向所述网络侧设备发送第三方链接,所述第三方链接用于获取所述用于确定质量评估方式的信息;
向所述网络侧设备发送应用程序编程接口API,所述应用程序编程接口API用于获取所述用于确定质量评估方式的信息。
可选地,在所述AI服务质量要求包括所述安全性要求的情况下,所述第一信息包括所述安全性要求指标门限和用于评估安全性质量的第一参数信息;其中,第一参数信息包括以下至少一项:
第一指示信息、第二信息、第一加权信息;
其中,第一指示信息用于指示舒适性质量评估包括的至少一个第一目标参数项;所述第一加权信息包括各个所述第一目标参数项的加权值;
所述第一目标参数项包括以下至少一项:
第一距离、第一保护间隔门限、第二保护间隔门限;
其中,第一距离是所述目标设备与障碍物的距离;所述第一保护间隔门限用于表
示最高安全等级的保护间隔下限,所述第二保护间隔门限用于表示最低安全等级的保护间隔下限;
第二信息为各个所述第一目标参数项与安全代价的函数映射信息,或者第一距离与安全等级的对应关系。
可选地,在所述第一距离位于第一保护间隔门限和第二保护间隔门限之间时,安全等级越高,所述第一距离越大。
可选地,所述第一距离为基于以下至少一项信息确定的;
所述至少一项信息包括:目标设备的尺寸、航向角、目标设备的位置和障碍物边界。
可选地,在所述AI服务质量要求包括所述舒适性要求的情况下,所述第一信息包括所述舒适性要求指标门限和用于评估舒适性质量的第二参数信息;其中,所述第二参数信息包括以下至少一项:
第二指示信息、第三信息、第二加权信息;
其中,第二指示信息用于指示舒适性质量评估包括的至少一个第二目标参数项,第三信息为各个所述第二目标参数项对应的舒适性代价函数的信息;所述第二加权信息包括各个所述第二目标参数项的加权值;所述第二目标参数项包括以下至少一项:
路径的朝向偏差、路径的航向角、路径的曲率、路径的曲率导数、路径的曲率变化率、加速度、加加速度。
可选地,在所述AI服务质量要求包括所述高效性要求的情况下,所述第一信息包括所述高效性要求指标门限和用于评估高效性质量的第三参数信息;其中,所述第三参数信息包括以下至少一项:
第三指示信息、第四信息、第三加权信息;
其中,第三指示信息用于指示高效性质量评估包括的至少一个第三目标参数项,第四信息为各个所述第三目标参数项对应的高效性代价函数的信息;所述第三加权信息包括各个所述第三目标参数项的加权值;所述第三目标参数项包括以下至少一项:
与车道中心线的偏差、与理想无障碍路径的偏差、超过道路边界的惩罚值、路径行驶时长、路径行驶距离。
可选地,所述AI服务质量要求还包括以下至少一种类型的要求:
速度保持代价;
路径单调性;
目标设备速度上限;
路径规划的总时长;
路径规划的时间间隔;
其中,速度保持代价为基于服务结果中包括的行驶速度与目标行驶速度的差值得到的,目标行驶速度为由第五信息确定的;所述第五信息包括以下至少一项:路径的曲率、交通规则;
路径单调性是指在一次规划的路径中所述目标设备与路径起点的距离是单调递增的,或所述目标设备与路径终点的距离是单调递减的;
目标设备速度上限为基于交通规则和/或目标设备动力学得到的。
可选地,所述第一信息中还包括:用于计算速度保持代价的信息,所述用于计算速度保持代价的信息包括以下至少一项:
偏离速度惩罚函数信息;
偏离速度的惩罚倍数。
可选地,所述AI服务质量要求还包括用于指示服务质量要求类型的指示信息,以及每个类型的服务质量要求的合并方式,所述合并方式包括以下至少一项:
每个类型的服务质量要求包括的参数项合并的加权函数信息;
每个类型的服务质量要求包括的参数项的加权值。
可选地,第一信息中还包括以下至少一项:
目标设备的尺寸;
目标设备的位置;
目标设备对周围物体的感知结果;所述感知结果包括以下至少一项的感知结果:类型、距离、速度、加速度、边界;
目标设备对周围物体的原始感知数据。
可选地,所述射频单元1001还用于:
接收所述网络侧设备发送的所述服务结果的第一评价结果;所述服务结果的第一评价结果为基于所述第六信息和所述服务结果确定的;所述第六信息包括以下至少一项:质量要求指标、质量要求指标计算方法、质量要求指标输入数据和格式、质量要求指标门限。
可选地,所述射频单元1001还用于:
向所述网络侧设备发送第二评价结果,所述第二评价结果为所述目标设备对所述服务结果进行评价的评价结果。
可选地,第一评价结果或第二评价结果包括以下至少一项:
AI服务标识;
AI服务标识对应的服务等级;
AI服务标识对应的服务质量数值;
综合服务质量数值;
综合服务质量等级。
可选地,综合服务质量要求为由所述第一信息确定的,包括以下至少一种方式:
基于所述第一信息直接得到的;
基于所述第一信息得到AI服务质量要求,再基于所述AI服务质量要求得到的。
可选地,在所述综合服务质量要求为由所述第一信息确定的情况下,所述第一信息还包括以下至少一项:
综合服务质量等级;
综合服务质量门限;
各服务质量的权重;
各服务质量的加权方法。
本申请实施例还提供一种网络侧设备,包括处理器和通信接口,所述通信接口用于接收来自目标设备的第一信息,所述第一信息包括第一服务的AI服务标识和用于指示AI服务质量要求的指示信息;所述AI服务质量要求包括以下至少一种类型的要求:安全性要求、舒适性要求、高效性要求和时延要求;所述安全性要求用于表征所述目标设备的防碰撞能力;所述舒适性要求用于表征所述目标设备的乘坐舒适度;所述高效性要求用于表征所述目标设备完成任务的效率;所述时延要求用于表征所述目标设备发起AI服务请求与获取到服务结果之间的时间差;所述处理器用于基于所述AI服务标识和AI服务质量要求,向目标设备反馈服务结果。该网络侧设备实施例与上述网
络侧设备方法实施例对应,上述方法实施例的各个实施过程和实现方式均可适用于该网络侧设备实施例中,且能达到相同的技术效果。
具体地,本申请实施例还提供了一种网络侧设备。如图9所示,该网络侧设备800包括:天线81、射频装置82、基带装置83、处理器85和存储器85。
天线81与射频装置82连接。
在上行方向上,射频装置82通过天线81接收信息,将接收的信息发送给基带装置83进行处理。
在下行方向上,基带装置83对要发送的信息进行处理,并发送给射频装置82,射频装置82对收到的信息进行处理后经过天线81发送出去。
上述频带处理装置可以位于基带装置83中,以上实施例中接入网设备执行的方法可以在基带装置83中实现,该基带装置83包括基带处理器85和存储器85。
基带装置83例如可以包括至少一个基带板,该基带板上设置有多个芯片,如图9所示,其中一个芯片例如为基带处理器85,通过总线接口与存储器85连接,以调用存储器85中的程序,执行以上方法实施例中所示的接入网设备操作。
接入网设备800还可以包括网络接口86,用于与射频装置82交互信息,该接口例如为通用公共无线接口(common public radio interface,简称CPRI)。
具体地,本申请实施的接入网设备800还包括:存储在存储器85上并可在处理器85上运行的指令或程序,处理器85调用存储器85中的指令或程序执行如图5所示模块执行的方法,并达到相同的技术效果,为避免重复,故不在此赘述。
本申请实施例还提供一种可读存储介质,所述可读存储介质上存储有程序或指令,该程序或指令被处理器执行时实现上述AI服务的处理方法实施例的各个过程,且能达到相同的技术效果,为避免重复,这里不再赘述。
其中,所述处理器为上述实施例中所述的终端中的处理器。所述可读存储介质,包括计算机可读存储介质,如计算机只读存储器ROM、随机存取存储器RAM、磁碟或者光盘等。
本申请实施例另提供了一种芯片,所述芯片包括处理器和通信接口,所述通信接口和所述处理器耦合,所述处理器用于运行程序或指令,实现上述AI服务的处理方法实施例的各个过程,且能达到相同的技术效果,为避免重复,这里不再赘述。
应理解,本申请实施例提到的芯片还可以称为系统级芯片,系统芯片,芯片系统或片上系统芯片等。
本申请实施例另提供了一种计算机程序/程序产品,所述计算机程序/程序产品被存储在存储介质中,所述计算机程序/程序产品被至少一个处理器执行以实现上述AI服务的处理方法实施例的各个过程,且能达到相同的技术效果,为避免重复,这里不再赘述。
本申请实施例还提供了一种通信系统,包括:目标设备及网络侧设备,所述目标设备可用于执行如上所述的AI服务的处理方法的步骤,所述网络侧设备可用于执行如上所述的AI服务的处理方法的步骤。
需要说明的是,在本文中,术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含,从而使得包括一系列要素的过程、方法、物品或者装置不仅包括那些要素,而且还包括没有明确列出的其他要素,或者是还包括为这种过程、方法、物品或者装置所固有的要素。在没有更多限制的情况下,由语句“包括一个……”限定的要素,并不排除在包括该要素的过程、方法、物品或者装置中还存在另外的相同
要素。此外,需要指出的是,本申请实施方式中的方法和装置的范围不限按示出或讨论的顺序来执行功能,还可包括根据所涉及的功能按基本同时的方式或按相反的顺序来执行功能,例如,可以按不同于所描述的次序来执行所描述的方法,并且还可以添加、省去、或组合各种步骤。另外,参照某些示例所描述的特征可在其他示例中被组合。
通过以上的实施方式的描述,本领域的技术人员可以清楚地了解到上述实施例方法可借助软件加必需的通用硬件平台的方式来实现,当然也可以通过硬件,但很多情况下前者是更佳的实施方式。基于这样的理解,本申请的技术方案本质上或者说对相关技术做出贡献的部分可以以计算机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介质(如ROM/RAM、磁碟、光盘)中,包括若干指令用以使得一台终端(可以是手机,计算机,服务器,空调器,或者网络设备等)执行本申请各个实施例所述的方法。
上面结合附图对本申请的实施例进行了描述,但是本申请并不局限于上述的具体实施方式,上述的具体实施方式仅仅是示意性的,而不是限制性的,本领域的普通技术人员在本申请的启示下,在不脱离本申请宗旨和权利要求所保护的范围情况下,还可做出很多形式,均属于本申请的保护之内。
Claims (30)
- 一种人工智能AI服务的处理方法,包括:网络侧设备接收来自目标设备的第一信息,所述第一信息包括第一服务的AI服务标识和用于指示AI服务质量要求的指示信息;所述AI服务质量要求包括以下至少一种类型的要求:安全性要求、舒适性要求、高效性要求和时延要求;所述安全性要求用于表征所述目标设备的防碰撞能力;所述舒适性要求用于表征所述目标设备的乘坐舒适度;所述高效性要求用于表征所述目标设备完成任务的效率;所述时延要求用于表征所述目标设备发起AI服务请求与获取到服务结果之间的时间差;所述网络侧设备基于所述AI服务标识和AI服务质量要求,向目标设备反馈所述第一服务的服务结果。
- 根据权利要求1所述的方法,其中,所述指示信息通过以下至少一项表示所述AI服务质量要求:质量等级、质量要求指标门限;其中,所述质量要求指标门限包括以下至少一项:安全性要求指标门限、舒适性要求指标门限、高效性要求指标门限和时延要求指标门限。
- 根据权利要求1或2所述的方法,其中,所述方法还包括以下至少一项:所述网络侧设备接收来自所述目标设备的用于确定质量评估方式的信息;所述网络侧设备接收来自所述目标设备的第三方链接,并基于所述第三方链接获取用于确定质量评估方式的信息;所述网络侧设备接收来自所述目标设备的应用程序编程接口API,并基于所述API获取用于确定质量评估方式的信息;其中,所述质量评估方式用于对服务结果的质量进行评估;所述用于确定质量评估方式的信息包括:质量评估方式对应的可执行文件和运行信息;所述运行信息包括以下至少一项:运行环境配置信息、文件名称、输入格式、输出格式、指示预处理方式的信息、指示后处理方式的信息。
- 根据权利要求1或2所述的方法,其中,在所述AI服务质量要求包括所述安全性要求的情况下,所述第一信息包括所述安全性要求指标门限和用于评估安全性质量的第一参数信息;其中,第一参数信息包括以下至少一项:第一指示信息、第二信息、第一加权信息;其中,第一指示信息用于指示安全性质量评估包括的至少一个第一目标参数项;所述第一加权信息包括各个所述第一目标参数项的加权值;所述第一目标参数项包括以下至少一项:第一距离、第一保护间隔门限、第二保护间隔门限;其中,第一距离是所述目标设备与障碍物的距离;所述第一保护间隔门限用于表示最高安全等级的保护间隔下限,所述第二保护间隔门限用于表示最低安全等级的保护间隔下限;第二信息为各个所述第一目标参数项与安全代价的函数映射信息,或者第一距离与安全等级的对应关系;所述第一距离为由以下至少一项信息确定的;所述至少一项信息包括:目标设备 的尺寸、航向角、目标设备的位置和障碍物边界。
- 根据权利要求1或2所述的方法,其中,在所述AI服务质量要求包括所述舒适性要求的情况下,所述第一信息包括所述舒适性要求指标门限和用于评估舒适性质量的第二参数信息;其中,所述第二参数信息包括以下至少一项:第二指示信息、第三信息、第二加权信息;其中,第二指示信息用于指示舒适性质量评估包括的至少一个第二目标参数项,第三信息为各个所述第二目标参数项对应的舒适性代价函数的信息;所述第二加权信息包括各个所述第二目标参数项的加权值;所述第二目标参数项包括以下至少一项:路径的朝向偏差、路径的航向角、路径的曲率、路径的曲率导数、路径的曲率变化率、加速度、加加速度。
- 根据权利要求1或2所述的方法,其中,在所述AI服务质量要求包括所述高效性要求的情况下,所述第一信息包括所述高效性要求指标门限和用于评估高效性质量的第三参数信息;其中,所述第三参数信息包括以下至少一项:第三指示信息、第四信息、第三加权信息;其中,第三指示信息用于指示高效性质量评估包括的至少一个第三目标参数项,第四信息为各个所述第三目标参数项对应的高效性代价函数的信息;所述第三加权信息包括各个所述第三目标参数项的加权值;所述第三目标参数项包括以下至少一项:与车道中心线的偏差、与理想无障碍路径的偏差、超过道路边界的惩罚值、路径行驶时长、路径行驶距离。
- 根据权利要求1或2所述的方法,其中,所述AI服务质量要求还包括以下至少一种类型的要求:速度保持代价;路径单调性;目标设备速度上限;路径规划的总时长;路径规划的时间间隔;其中,速度保持代价为基于服务结果中包括的行驶速度与目标行驶速度的差值得到的,目标行驶速度为由第五信息确定的;所述第五信息包括以下至少一项:路径的曲率、交通规则;路径单调性是指在一次规划的路径中所述目标设备与路径起点的距离是单调递增的,或所述目标设备与路径终点的距离是单调递减的;目标设备速度上限为基于交通规则和/或目标设备动力学得到的;所述第一信息中还包括:用于计算速度保持代价的信息,所述用于计算速度保持代价的信息包括以下至少一项:偏离速度惩罚函数信息;偏离速度的惩罚倍数。
- 根据权利要求4-6任一项所述的方法,其中,所述AI服务质量要求还包括:用于指示服务质量要求类型的指示信息,以及每个类型的服务质量要求的合并方式,所述合并方式包括以下至少一项:每个类型的服务质量要求包括的参数项合并的加权函数信息;每个类型的服务质量要求包括的参数项的加权值。
- 根据权利要求1或2所述的方法,其中,第一信息中还包括以下至少一项:目标设备的尺寸;目标设备的位置;目标设备对周围物体的感知结果;所述感知结果包括以下至少一项的感知结果:类型、距离、速度、加速度、边界;目标设备对周围物体的原始感知数据。
- 根据权利要求1-3任一项所述的方法,其中,所述方法还包括:所述网络侧设备基于所述第一信息和质量评估方式的信息,或,基于所述第一信息确定第六信息,所述第六信息包括以下至少一项:质量要求指标、质量要求指标计算方法、质量要求指标输入数据和格式、质量要求指标门限。
- 根据权利要求10所述的方法,其中,所述方法还包括:所述网络侧设备基于所述第六信息和所述服务结果,确定所述服务结果的第一评价结果;所述网络侧设备向所述目标设备发送所述服务结果的第一评价结果;所述第一评价结果包括以下至少一项:AI服务标识;AI服务标识对应的服务等级;AI服务标识对应的服务质量数值;综合服务质量数值;综合服务质量等级。
- 根据权利要求1或2所述的方法,其中,所述方法还包括:所述网络侧设备接收所述目标设备发送的第二评价结果,所述第二评价结果为所述目标设备对所述服务结果进行评价的评价结果;所述第二评价结果包括以下至少一项:AI服务标识;AI服务标识对应的服务等级;AI服务标识对应的服务质量数值;综合服务质量数值;综合服务质量等级。
- 根据权利要求1-3任一项所述的方法,其中,所述网络侧设备基于所述AI服务标识和AI服务质量要求,向目标设备反馈服务结果,包括:所述网络侧设备基于所述AI服务标识和第七信息,确定提供服务的算法信息;所述第七信息包括以下至少一项:AI服务质量要求、综合服务质量要求;所述网络侧设备基于所述AI服务标识以及所述算法信息,向目标设备反馈所述服务结果。
- 根据权利要求13所述的方法,其中,所述综合服务质量要求为由所述第一信息确定的,包括以下至少一种方式:基于所述第一信息直接得到的;基于所述第一信息得到AI服务质量要求,再基于所述AI服务质量要求得到的;所述第一信息还包括以下至少一项:综合服务质量等级;综合服务质量门限;各服务质量的权重;各服务质量的加权方法。
- 一种AI服务的处理方法,包括:目标设备向网络侧设备发送第一信息,所述第一信息包括第一服务的AI服务标识和用于指示AI服务质量要求的指示信息;所述AI服务质量要求包括以下至少一种类型的要求:安全性要求、舒适性要求、高效性要求和时延要求;所述安全性要求用于表征所述目标设备的防碰撞能力;所述舒适性要求用于表征所述目标设备的乘坐舒适度;所述高效性要求用于表征所述目标设备完成任务的效率;所述时延要求用于表征所述目标设备发起AI服务请求与获取到服务结果之间的时间差;所述目标设备接收所述网络侧设备发送的所述第一服务的服务结果。
- 根据权利要求15所述的方法,其中,所述指示信息通过以下至少一项表示所述AI服务质量要求:质量等级、质量要求指标门限;其中,所述质量要求指标门限包括以下至少一项:安全性要求指标门限、舒适性要求指标门限、高效性要求指标门限和时延要求指标门限。
- 根据权利要求15或16所述的方法,其中,所述方法还包括以下至少一项:所述目标设备向所述网络侧设备发送用于确定质量评估方式的信息;所述目标设备向所述网络侧设备发送第三方链接,所述第三方链接用于获取用于确定质量评估方式的信息;所述目标设备向所述网络侧设备发送应用程序编程接口API,所述应用程序编程接口API用于获取用于确定质量评估方式的信息;其中,所述质量评估方式用于对服务结果的质量进行评估;所述用于确定质量评估方式的信息包括:质量评估方式对应的可执行文件和运行信息;所述运行信息包括以下至少一项:运行环境配置信息、文件名称、输入格式、输出格式、指示预处理方式的信息、指示后处理方式的信息。
- 根据权利要求15或16所述的方法,其中,在所述AI服务质量要求包括所述安全性要求的情况下,所述第一信息包括所述安全性要求指标门限和用于评估安全性质量的第一参数信息;其中,第一参数信息包括以下至少一项:第一指示信息、第二信息、第一加权信息;其中,第一指示信息用于指示安全性质量评估包括的至少一个第一目标参数项;所述第一加权信息包括各个所述第一目标参数项的加权值;所述第一目标参数项包括以下至少一项:第一距离、第一保护间隔门限、第二保护间隔门限;其中,第一距离是所述目标设备与障碍物的距离;所述第一保护间隔门限用于表示最高安全等级的保护间隔下限,所述第二保护间隔门限用于表示最低安全等级的保护间隔下限;第二信息为各个所述第一目标参数项与安全代价的函数映射信息,或者第一距离与安全等级的对应关系;所述第一距离为基于以下至少一项信息确定的;所述至少一项信息包括:目标设备的尺寸、航向角、目标设备的位置和障碍物边界。
- 根据权利要求15或16所述的方法,其中,在所述AI服务质量要求包括所述舒适性要求的情况下,所述第一信息包括所述舒适性要求指标门限和用于评估舒适性质量的第二参数信息;其中,所述第二参数信息包括以下至少一项:第二指示信息、第三信息、第二加权信息;其中,第二指示信息用于指示舒适性质量评估包括的至少一个第二目标参数项,第三信息为各个所述第二目标参数项对应的舒适性代价函数的信息;所述第二加权信息包括各个所述第二目标参数项的加权值;所述第二目标参数项包括以下至少一项:路径的朝向偏差、路径的航向角、路径的曲率、路径的曲率导数、路径的曲率变化率、加速度、加加速度。
- 根据权利要求15或16所述的方法,其中,在所述AI服务质量要求包括所述高效性要求的情况下,所述第一信息包括所述高效性要求指标门限和用于评估高效性质量的第三参数信息;其中,所述第三参数信息包括以下至少一项:第三指示信息、第四信息、第三加权信息;其中,第三指示信息用于指示高效性质量评估包括的至少一个第三目标参数项,第四信息为各个所述第三目标参数项对应的高效性代价函数的信息;所述第三加权信息包括各个所述第三目标参数项的加权值;所述第三目标参数项包括以下至少一项:与车道中心线的偏差、与理想无障碍路径的偏差、超过道路边界的惩罚值、路径行驶时长、路径行驶距离。
- 根据权利要求15或16所述的方法,其中,所述AI服务质量要求还包括以下至少一种类型的要求:速度保持代价;路径单调性;目标设备速度上限;路径规划的总时长;路径规划的时间间隔;其中,速度保持代价为基于服务结果中包括的行驶速度与目标行驶速度的差值得到的,目标行驶速度为由第五信息确定的;所述第五信息包括以下至少一项:路径的曲率、交通规则;路径单调性是指在一次规划的路径中所述目标设备与路径起点的距离是单调递增的,或所述目标设备与路径终点的距离是单调递减的;目标设备速度上限为基于交通规则和/或目标设备动力学得到的;所述第一信息中还包括:用于计算速度保持代价的信息,所述用于计算速度保持代价的信息包括以下至少一项:偏离速度惩罚函数信息;偏离速度的惩罚倍数。
- 根据权利要求18-20任一项所述的方法,其中,所述AI服务质量要求还包括用于指示服务质量要求类型的指示信息,以及每个类型的服务质量要求的合并方式,所述合并方式包括以下至少一项:每个类型的服务质量要求包括的参数项合并的加权函数信息;每个类型的服务质量要求包括的参数项的加权值。
- 根据权利要求15或16所述的方法,其中,所述方法还包括:所述目标设备接收所述网络侧设备发送的所述服务结果的第一评价结果;所述服务结果的第一评价结果为基于第六信息和所述服务结果确定的;所述第六信息包括以下至少一项:质量要求指标、质量要求指标计算方法、质量要求指标输入数据和格式、质量要求指标门限;所述第一评价结果包括以下至少一项:AI服务标识;AI服务标识对应的服务等级;AI服务标识对应的服务质量数值;综合服务质量数值;综合服务质量等级。
- 根据权利要求15或16所述的方法,其中,所述方法还包括:所述目标设备向所述网络侧设备发送第二评价结果,所述第二评价结果为所述目标设备对所述服务结果进行评价的评价结果;所述第二评价结果包括以下至少一项:AI服务标识;AI服务标识对应的服务等级;AI服务标识对应的服务质量数值;综合服务质量数值;综合服务质量等级。
- 根据权利要求15或16所述的方法,其中,综合服务质量要求为由所述第一信息确定的,包括以下至少一种方式:基于所述第一信息直接得到的;基于所述第一信息得到AI服务质量要求,再基于所述AI服务质量要求得到的;所述第一信息还包括以下至少一项:综合服务质量等级;综合服务质量门限;各服务质量的权重;各服务质量的加权方法。
- 一种AI服务的处理装置,包括:接收模块,用于接收来自目标设备的第一信息,所述第一信息包括第一服务的AI服务标识和用于指示AI服务质量要求的指示信息;所述AI服务质量要求包括以下至少一种类型的要求:安全性要求、舒适性要求、高效性要求和时延要求;所述安全性要求用于表征评估所述目标设备的防碰撞能力;所述舒适性要求用于表征所述目标设备的乘坐舒适度;所述高效性要求用于表征所述目标设备完成任务的效率;所述时延要求用于表征所述目标设备发起AI服务请求与获取到服务结果之间的时间差;处理模块,用于基于所述AI服务标识和AI服务质量要求,向目标设备反馈第一服务的服务结果。
- 一种AI服务的处理装置,包括:发送模块,用于向网络侧设备发送第一信息,所述第一信息包括第一服务的AI服务标识和用于指示AI服务质量要求的指示信息;所述AI服务质量要求包括以下至 少一种类型的要求:安全性要求、舒适性要求、高效性要求和时延要求;所述安全性要求用于表征目标设备的防碰撞能力;所述舒适性要求用于表征目标设备的乘坐舒适度;所述高效性要求用于表征目标设备完成任务的效率;所述时延要求用于表征所述目标设备发起AI服务请求与获取到服务结果之间的时间差;接收模块,用于接收所述网络侧设备发送的第一服务的服务结果。
- 一种网络侧设备,包括处理器和存储器,所述存储器存储可在所述处理器上运行的程序或指令,所述程序或指令被所述处理器执行时实现如权利要求1至14任一项所述的AI服务的处理方法的步骤。
- 一种目标设备,包括处理器和存储器,所述存储器存储可在所述处理器上运行的程序或指令,所述程序或指令被所述处理器执行时实现如权利要求15至25任一项所述的AI服务的处理方法的步骤。
- 一种可读存储介质,所述可读存储介质上存储程序或指令,所述程序或指令被处理器执行时实现如权利要求1至14任一项所述的AI服务的处理方法,或者实现如权利要求15至25任一项所述的AI服务的处理方法的步骤。
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211387610.2A CN117998325A (zh) | 2022-11-07 | 2022-11-07 | 人工智能ai服务的处理方法及设备 |
CN202211387610.2 | 2022-11-07 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2024099111A1 true WO2024099111A1 (zh) | 2024-05-16 |
Family
ID=90895064
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2023/127307 WO2024099111A1 (zh) | 2022-11-07 | 2023-10-27 | Ai服务的处理方法及设备 |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN117998325A (zh) |
WO (1) | WO2024099111A1 (zh) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112953778A (zh) * | 2021-03-17 | 2021-06-11 | 北京交通大学 | 智融标识网络中基于意图驱动的服务编排系统和方法 |
CN114026834A (zh) * | 2019-04-30 | 2022-02-08 | 英特尔公司 | 边缘计算部署中的多实体资源、安全性和服务管理 |
CN114675945A (zh) * | 2021-05-21 | 2022-06-28 | 腾讯云计算(北京)有限责任公司 | 一种服务调用方法、装置、计算机设备及存储介质 |
CN114745264A (zh) * | 2020-12-23 | 2022-07-12 | 大唐移动通信设备有限公司 | 推理服务的部署方法、装置及处理器可读存储介质 |
-
2022
- 2022-11-07 CN CN202211387610.2A patent/CN117998325A/zh active Pending
-
2023
- 2023-10-27 WO PCT/CN2023/127307 patent/WO2024099111A1/zh unknown
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114026834A (zh) * | 2019-04-30 | 2022-02-08 | 英特尔公司 | 边缘计算部署中的多实体资源、安全性和服务管理 |
CN114745264A (zh) * | 2020-12-23 | 2022-07-12 | 大唐移动通信设备有限公司 | 推理服务的部署方法、装置及处理器可读存储介质 |
CN112953778A (zh) * | 2021-03-17 | 2021-06-11 | 北京交通大学 | 智融标识网络中基于意图驱动的服务编排系统和方法 |
CN114675945A (zh) * | 2021-05-21 | 2022-06-28 | 腾讯云计算(北京)有限责任公司 | 一种服务调用方法、装置、计算机设备及存储介质 |
Also Published As
Publication number | Publication date |
---|---|
CN117998325A (zh) | 2024-05-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN114071586A (zh) | 端到端QoS需求信息的拆分方法、终端及网络侧设备 | |
WO2024099111A1 (zh) | Ai服务的处理方法及设备 | |
EP3674893B1 (en) | Search resource recommendation method and related product | |
CN111278114A (zh) | 功率控制方法、终端设备及网络侧设备 | |
CN116193500A (zh) | 感知业务实现方法、装置、网络侧设备及终端 | |
WO2024149288A1 (zh) | Ai模型分发、接收方法、终端及网络侧设备 | |
WO2024099140A1 (zh) | 人工智能ai服务的处理方法及设备 | |
WO2024153013A1 (zh) | 信息传输方法、装置及通信设备 | |
WO2024088119A1 (zh) | 数据处理方法、装置、终端及网络侧设备 | |
WO2023213242A1 (zh) | Ue策略信息的配置方法、装置、第一网络功能及终端 | |
WO2024120444A1 (zh) | 模型监督方法、装置、终端、网络侧设备及可读存储介质 | |
WO2024067438A1 (zh) | Ai模型推理方法、设备及可读存储介质 | |
WO2024208260A1 (zh) | 基于ai的csi预测的数据采集方法及装置 | |
WO2024138375A1 (zh) | 一种通信方法、装置、设备及存储介质 | |
US20240188046A1 (en) | Computing session update method and apparatus, terminal, and network side device | |
WO2023246756A1 (zh) | 算力服务方法、装置、终端及核心网设备 | |
WO2024067280A1 (zh) | 更新ai模型参数的方法、装置及通信设备 | |
WO2024067436A1 (zh) | 信息传输方法、装置及设备 | |
WO2023165393A1 (zh) | 传输方法、装置、终端和存储介质 | |
WO2023186099A1 (zh) | 信息反馈方法、装置及设备 | |
WO2024022398A1 (zh) | 托管网络的选网信息的获取方法、终端及网络侧设备 | |
CN116489035A (zh) | 通信网络预测方法、终端及网络侧设备 | |
CN117440449A (zh) | Ai模型信息传输方法、装置及设备 | |
CN118265050A (zh) | 策略配置方法、装置、终端、网络侧设备及可读存储介质 | |
CN116866886A (zh) | 数据获取方法及通信设备 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 23887801 Country of ref document: EP Kind code of ref document: A1 |