WO2020037500A1 - 智能驾驶方法、装置及存储介质 - Google Patents

智能驾驶方法、装置及存储介质 Download PDF

Info

Publication number
WO2020037500A1
WO2020037500A1 PCT/CN2018/101568 CN2018101568W WO2020037500A1 WO 2020037500 A1 WO2020037500 A1 WO 2020037500A1 CN 2018101568 W CN2018101568 W CN 2018101568W WO 2020037500 A1 WO2020037500 A1 WO 2020037500A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
decision
road condition
artificial intelligence
artificial
Prior art date
Application number
PCT/CN2018/101568
Other languages
English (en)
French (fr)
Inventor
廉士国
刘兆祥
Original Assignee
深圳前海达闼云端智能科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳前海达闼云端智能科技有限公司 filed Critical 深圳前海达闼云端智能科技有限公司
Priority to CN201880001564.3A priority Critical patent/CN109196437B/zh
Priority to PCT/CN2018/101568 priority patent/WO2020037500A1/zh
Publication of WO2020037500A1 publication Critical patent/WO2020037500A1/zh

Links

Classifications

    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/02Control of position or course in two dimensions
    • G05D1/021Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0231Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means
    • G05D1/0246Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using a video camera in combination with image processing means
    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/02Control of position or course in two dimensions
    • G05D1/021Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0212Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
    • G05D1/0214Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory in accordance with safety or protection criteria, e.g. avoiding hazardous areas
    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/02Control of position or course in two dimensions
    • G05D1/021Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0212Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
    • G05D1/0223Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving speed control of the vehicle
    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/02Control of position or course in two dimensions
    • G05D1/021Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0255Control of position or course in two dimensions specially adapted to land vehicles using acoustic signals, e.g. ultra-sonic singals
    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/02Control of position or course in two dimensions
    • G05D1/021Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0257Control of position or course in two dimensions specially adapted to land vehicles using a radar
    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/02Control of position or course in two dimensions
    • G05D1/021Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0276Control of position or course in two dimensions specially adapted to land vehicles using signals provided by a source external to the vehicle
    • G05D1/0278Control of position or course in two dimensions specially adapted to land vehicles using signals provided by a source external to the vehicle using satellite positioning signals, e.g. GPS
    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/02Control of position or course in two dimensions
    • G05D1/021Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0276Control of position or course in two dimensions specially adapted to land vehicles using signals provided by a source external to the vehicle
    • G05D1/0285Control of position or course in two dimensions specially adapted to land vehicles using signals provided by a source external to the vehicle using signals transmitted via a public communication network, e.g. GSM network

Definitions

  • the present disclosure relates to the field of artificial intelligence technologies, and in particular, to a method, device, and storage medium for intelligent driving.
  • AI artificial intelligence
  • the present disclosure provides a smart driving method, device, and storage medium.
  • a first aspect of the present disclosure provides an intelligent driving method, including:
  • the artificial intelligence perception model is configured to output road condition information according to sensor information, and the first confidence level indicates that the artificial intelligence perception model outputs correct road condition information The probability;
  • the artificial intelligence decision model is configured to output decision information according to the target road condition information, and the second confidence level indicates that the artificial intelligence decision model output is correct Probability of decision information;
  • a second aspect of the present disclosure provides an intelligent driving device, including:
  • a first determining module configured to determine a first confidence level of a preset artificial intelligence perception model, the artificial intelligence perception model is configured to output road condition information according to sensor information, and the first confidence level represents the artificial intelligence Probability that the perception model outputs correct road condition information;
  • a first acquisition module configured to acquire, as the target road condition information, road condition information output by the artificial intelligence sensing model according to the sensor information when the first confidence level is greater than or equal to a first preset threshold
  • a second determining module configured to determine a second confidence level of the preset artificial intelligence decision model on the target road condition information, the artificial intelligence decision model used to output decision information according to the target road condition information, and the second The confidence degree indicates the probability that the artificial intelligence decision model outputs correct decision information
  • a first display module configured to display the target road condition information if the second confidence level is less than a second preset threshold
  • a second acquisition module configured to use the received first artificial decision information on the target road condition information as target decision information, and the target decision information is used to control the vehicle.
  • a third aspect of the present disclosure provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the steps of the method described in the first aspect of the present disclosure.
  • a fourth aspect of the present disclosure provides an intelligent driving device, including: the computer-readable storage medium according to the third aspect of the present disclosure; and one or more processors configured to execute the computer-readable storage medium. program of.
  • Determining whether to trigger a manual intervention decision based on the confidence of a preset artificial intelligence perception model and the confidence of a preset artificial intelligence decision model can enable adaptive triggering when it is difficult to make a reliable decision through artificial intelligence.
  • Manual intervention in decision-making to ensure reliable decision-making results, and controlling vehicles based on reliable decision-making results can improve vehicle driving safety.
  • FIG. 1 is a flowchart illustrating a smart driving method according to an exemplary embodiment of the present disclosure
  • FIG. 2 is a flowchart illustrating a smart driving method according to another exemplary embodiment of the present disclosure
  • FIG. 3 is a block diagram of a smart driving device according to an exemplary embodiment of the present disclosure.
  • FIG. 4 is a block diagram of a smart driving device according to another exemplary embodiment of the present disclosure.
  • FIG. 5 is a block diagram of an intelligent driving system according to an exemplary embodiment of the present disclosure.
  • FIG. 6 is a block diagram of a smart driving device according to another exemplary embodiment of the present disclosure.
  • FIG. 1 is a flowchart illustrating a smart driving method according to an exemplary embodiment of the present disclosure. As shown in FIG. 1, the method includes the following steps:
  • step S11 a first confidence level of a preset artificial intelligence sensing model is determined, and the artificial intelligence sensing model is used to output road condition information according to sensor information.
  • the first confidence level indicates the probability that the artificial intelligence perception model outputs correct road condition information, where the higher the first confidence level, it indicates the probability that the artificial intelligence perception model outputs correct road condition information.
  • the larger the value the higher the reliability of the road condition information output by the artificial intelligence perception model.
  • the artificial intelligence perception model may be a Convolutional Neural Network (Convolutional Neural Network, CNN) model.
  • CNN Convolutional Neural Network
  • the sensor information may be collected by a collection device provided at an appropriate position on the vehicle, where the collection device may include, but is not limited to, a radar, a millimeter wave sensor, an ultrasonic sensor, a GPS locator, and image acquisition. And so on. Accordingly, the sensor information may include, but is not limited to, information such as sound, distance, and image.
  • the road condition information output by the artificial intelligence sensing model according to the sensor information is obtained as the target road condition information.
  • the first confidence level is greater than or equal to the first preset threshold, it can be considered that the reliability of the road condition information output by the artificial intelligence sensing model according to the sensor information is high, so artificial intelligence can be triggered to sense, that is, to obtain artificial intelligence perception
  • the model uses the road condition information output by the sensor information as the target road condition information.
  • the road condition information may include, but is not limited to, information about pedestrians around the vehicle, remaining vehicles, lane lines, and traffic signs.
  • step S13 a second confidence level of the preset artificial intelligence decision model is determined, and the artificial intelligence decision model is used to output decision information according to the target road condition information.
  • the second confidence degree indicates a probability that the artificial intelligence decision model outputs correct decision information according to the target road condition information, wherein the higher the second confidence degree indicates that the artificial intelligence decision model outputs correct decisions
  • the greater the probability of the information the more reliable the decision information output by the artificial intelligence decision model can be considered.
  • the artificial intelligence decision model may be a recurrent neural network (Recurrent Neural
  • step S14 if the second confidence level is less than a second preset threshold, target road condition information is displayed.
  • step S15 the received first artificial decision information on the target road condition information is used as the target decision information, and the target decision information is used to control the vehicle.
  • the second confidence level is less than the second preset threshold, it can be considered that the reliability of the decision information output by the artificial intelligence decision model according to the target road condition information is low.
  • manual intervention can be triggered at this time.
  • Decision making that is, displaying target road condition information to a target user, and the target user makes a decision according to the target road condition information and inputs corresponding decision information (that is, first manual decision information).
  • the first artificial decision information may be used as the target decision information, and the vehicle may be controlled according to the target decision information, such as controlling vehicle acceleration, deceleration, and turning.
  • the first decision information may include, but is not limited to, acceleration, speed, driving direction, and the like of the vehicle.
  • the target user may include, but is not limited to, a driver and a passenger, and a customer service staff who remotely monitors the vehicle. , Control staff and so on.
  • the target road condition information may be displayed in a manner of constructing an abstracted image.
  • road condition parameters such as, but not limited to, the relative position relationship between obstacles around the vehicle and the vehicle, lane lines in front of the vehicle, traffic signs, etc.
  • the destination For example, a vehicle display device, a remote monitoring device, a mobile phone carried by the target user, and a helmet worn, etc.
  • the destination end reconstructs a road condition image according to the road condition parameters and displays the road condition image to the target user, so that the target user makes a decision based on the road condition image .
  • the bandwidth can be reduced and the transmission rate can be increased.
  • determining whether to trigger an artificial intervention decision based on a preset confidence level of an artificial intelligence perception model and a preset confidence level of an artificial intelligence decision model can make it difficult to make a reliable decision through artificial intelligence.
  • manual intervention decision-making is adaptively triggered to ensure that a reliable decision result is obtained, and controlling the vehicle based on the reliable decision result can improve vehicle driving safety.
  • the artificial intelligence decision model may also be updated by using the target road condition information and the first artificial decision information, so that Optimize and improve the performance of this artificial intelligence decision model, so that the next time that the same or similar target road condition information is obtained, the reliability of the decision information output by the artificial intelligence decision model according to the target road condition information is further reduced, and manual intervention is further reduced to save manpower.
  • the target road condition information and the first artificial decision information can be used as a data sample pair to train an artificial intelligence decision model.
  • the second confidence level is greater than or equal to the second preset threshold, it can be considered that the reliability of the decision information output by the artificial intelligence decision model is high, and at this time, Triggering an artificial intelligence decision, that is, obtaining decision information output by an artificial intelligence decision model according to target road condition information as target decision information, and using the target decision information to control a vehicle.
  • the first confidence level when the first confidence level is less than the first preset threshold, it can be considered that the reliability of the road condition information output by the artificial intelligence sensing model is low, and at this time, artificial Involved in perception and decision-making, that is, the sensor information can be displayed to the target user, and the target user judges the road condition based on the collected sensor information and makes a decision based on the judgment result, and inputs the judgment result (artificial perception information) and decision information (that is, the second artificial decision) Information).
  • artificial Involved in perception and decision-making that is, the sensor information can be displayed to the target user, and the target user judges the road condition based on the collected sensor information and makes a decision based on the judgment result, and inputs the judgment result (artificial perception information) and decision information (that is, the second artificial decision) Information).
  • the second artificial decision information may be used as the target decision information, and according to the The target decision information controls the vehicle, for example, controlling vehicle acceleration, deceleration, turning, and so on.
  • the second decision information may include, but is not limited to, the speed, acceleration, and driving direction of the vehicle.
  • the target user may include, but is not limited to, a driver and a passenger, a customer service staff who remotely monitors the vehicle, a control staff, and the like.
  • the manner of displaying the sensor information may include Augmented Reality (AR), Virtual Reality (VR), and Mixed Reality (MR) And other methods to assist target user perception and decision-making.
  • AR Augmented Reality
  • VR Virtual Reality
  • MR Mixed Reality
  • the artificial intelligence perception model may also be updated by using the sensor information and the received artificial perception information, and The artificial intelligence decision information and the second artificial decision information are used to update the artificial intelligence decision model.
  • sensor information and artificial perception information can be labeled as data sample pairs to train an artificial intelligence perception model; for an artificial intelligence decision model, artificial perception information and second artificial decision information can be used as data sample pairs. To train artificial intelligence decision models.
  • artificially intervening data to update artificial intelligence perception models and artificial intelligence decision models can optimize and enhance the performance of artificial intelligence to improve the reliability of artificial intelligence perception results and artificial intelligence decision results, and further ensure the safety of vehicle driving.
  • manual intervention can be further reduced to achieve the goal of saving manpower.
  • the intelligent driving method shown in the above embodiments of the present disclosure may be preferably applied to a cloud server.
  • FIG. 3 is a block diagram of an intelligent driving device according to an exemplary embodiment of the present disclosure. As shown in FIG. 3, the device 300 may include:
  • a first determination module 301 configured to determine a first confidence level of a preset artificial intelligence perception model, the artificial intelligence perception model is configured to output road condition information according to sensor information, and the first confidence level indicates the artificial Probability of intelligent perception model outputting correct road condition information;
  • a first obtaining module 302 configured to obtain, as the target road condition information, road condition information output by the artificial intelligence sensing model according to the sensor information when the first confidence level is greater than or equal to a first preset threshold;
  • a second determining module 303 for determining a second confidence level of a preset artificial intelligence decision model, the artificial intelligence decision model for outputting decision information according to the target road condition information, and the second confidence level representing Describe the probability that the artificial intelligence decision model outputs correct decision information;
  • a first display module 304 configured to display the target road condition information if the second confidence level is less than a second preset threshold
  • the second acquisition module 305 is configured to use the received first artificial decision information on the target road condition information as target decision information, and the target decision information is used to control the vehicle.
  • the device 300 further includes:
  • a first update module 306 configured to update the artificial intelligence decision model using the target road condition information and the first artificial decision information.
  • the device 300 further includes:
  • a third acquisition module 307 configured to acquire, if the second confidence level is greater than or equal to the second preset threshold, the decision information output by the artificial intelligence decision model according to the target road condition information as a target decision information.
  • the device 300 further includes:
  • a second display module 308 configured to display the sensor information when the first confidence level is less than the first preset threshold
  • a fourth acquisition module 309 is configured to receive artificial perception information and second artificial decision information on the sensor information, and use the second artificial decision information as target decision information.
  • the device 300 further includes:
  • a second update module 310 configured to update the artificial intelligence perception model by using the sensor information and the artificial perception information
  • a third update module 311 is configured to update the artificial intelligence decision model using the artificial perception information and the second artificial decision information.
  • the first display module 304 includes: [0074] a first determining submodule 341 is configured to determine a road condition parameter according to the target road condition information;
  • a first display sub-module 342 is configured to send the road condition parameter to a destination, to instruct the destination to generate and display a road condition image according to the road condition parameter.
  • the manner of displaying the sensor information includes: augmented reality, virtual reality, and mixed reality.
  • artificially intervening data to update artificial intelligence perception models and artificial intelligence decision models can optimize and enhance the performance of artificial intelligence to improve the reliability of artificial intelligence perception results and artificial intelligence decision results, and further ensure the safety of vehicle driving.
  • manual intervention can be further reduced to achieve the goal of saving manpower.
  • An embodiment of the present disclosure further provides a vehicle, and the vehicle may include an information collection device and an intelligent driving device provided by the foregoing embodiment of the present disclosure.
  • an embodiment of the present disclosure also provides an intelligent driving system.
  • the intelligent driving system may include: a cloud server 500, a display device 600, and a vehicle control device 700 located at the scene.
  • the cloud server 500 includes a smart driving device 510.
  • the display device 600 may be, for example, a display screen, a virtual reality helmet, an augmented reality helmet, or the like.
  • the vehicle control device 700 may be a terminal such as a vehicle, a mobile phone, or a computer, and may include an information acquisition device 710 and a decision execution device 720.
  • the information acquisition device 710 may be, for example, an image acquisition device, an ultrasonic sensor, a millimeter wave sensor, or a GPS positioner. , Radar, etc., which can be used to collect sensor information about the surrounding environment of the vehicle.
  • the sensor information may include, for example, but not limited to, information such as sound, image, and distance.
  • the information collection device 710 may send the collected sensor information to the cloud service in a wired (for example, cable, power grid) or wireless (for example, Bluetooth, WIFI) manner.
  • Device 500, the intelligent driving device 510 of the cloud server 500 analyzes the sensor information to obtain road condition information and makes a decision based on the road condition information, and sends the decision information to the decision execution device 720 of the vehicle control device 700, and the decision execution device 720 according to the decision information Control the vehicle.
  • the road condition information may include, but is not limited to, information about pedestrians around the vehicle, other vehicles, lane lines, and traffic signs.
  • the decision information may include, for example, information for controlling vehicle acceleration, deceleration, and turning.
  • the intelligent driving device 510 may include processors 522, the number of which may be one or more, and a memory 532, for storing a computer program executable by the processor 522.
  • the computer program stored in the memory 532 may include one or more modules each corresponding to a set of instructions.
  • the processor 522 may be configured to execute the computer program to perform the above-mentioned intelligent driving method.
  • the intelligent driving device 510 may further include a power supply component 526 and a communication component 550.
  • the power supply component 526 may be configured to perform power management of the intelligent driving device 510
  • the communication component 550 may be configured to implement the intelligent driving device 510 communication, for example, wired or wireless communication.
  • the intelligent driving device 510 may further include an input / output (I / O) interface 558.
  • the intelligent driving device 510 can operate based on an operating system stored in the memory 532, such as Windows ServerTM, Mac OS
  • a computer-readable storage medium including program instructions is provided, and the program instructions implement the steps of the above-mentioned intelligent driving method when executed by a processor.
  • the computer-readable storage medium may be the above-mentioned memory 532 including program instructions, and the above-mentioned program instructions may be executed by the processor 522 of the intelligent driving device 510 to complete the above-mentioned intelligent driving method.

Landscapes

  • Engineering & Computer Science (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Physics & Mathematics (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Multimedia (AREA)
  • Electromagnetism (AREA)
  • Acoustics & Sound (AREA)
  • Traffic Control Systems (AREA)

Abstract

一种智能驾驶方法、装置及存储介质,所述方法优选地应用于云端服务器,包括:确定人工智能感知模型的第一置信度,人工智能感知模型用于根据传感器信息输出路况信息(S11);当第一置信度大于或等于第一预设阈值时,获取人工智能感知模型根据传感器信息输出的路况信息作为目标路况信息(S12);确定人工智能决策模型的第二置信度,人工智能决策模型用于根据目标路况信息输出决策信息(S13);若第二置信度小于第二预设阈值,则显示目标路况信息(S14);将接收到的对目标路况信息的第一人工决策信息作为目标决策信息,目标决策信息用于对车辆进行控制(S15)。采用该技术方案,可以在通过人工智能难以做出可靠决策时,自适应地触发人工介入决策来确保得到可靠的决策结果。

Description

发明名称:智能驾驶方法、 装置及存储介质
技术领域
[0001] 本公开涉及人工智能技术领域, 尤其涉及一种智能驾驶方法、 装置及存储介质 背景技术
[0002] 随着科技的高速发展, 人工智能 (Artificial Intelligence, AI) 技术应运而生, 其在无人驾驶、 导盲机器人、 巡逻机器人等领域得到了广泛应用。
[0003] 尽管人工智能技术已经展示出了强大的能力, 但在某些方面人仍然存在不足, 例如在无人驾驶领域, 由于车辆行驶环境的复杂性, 目前的人工智能很难保证 1 00%的智能能力, 使得仅依靠人能智能技术对车辆进行控制很难确保车辆行驶的 安全性。
[0004] 发明内容
[0005] 为了克服现有技术中存在的问题, 本公开提供一种智能驾驶方法、 装置及存储 介质。
[0006] 为了实现上述目的, 本公开第一方面提供一种智能驾驶方法, 包括:
[0007] 确定预设的人工智能感知模型的第一置信度, 所述人工智能感知模型用于根据 传感器信息输出路况信息, 所述第一置信度表示所述人工智能感知模型输出正 确的路况信息的概率;
[0008] 当所述第一置信度大于或等于第一预设阈值时, 获取所述人工智能感知模型根 据所述传感器信息输出的路况信息作为目标路况信息;
[0009] 确定预设的人工智能决策模型的第二置信度, 所述人工智能决策模型用于根据 所述目标路况信息输出决策信息, 所述第二置信度表示所述人工智能决策模型 输出正确的决策信息的概率;
[0010] 若所述第二置信度小于第二预设阈值, 则显示所述目标路况信息;
[0011] 将接收到的对所述目标路况信息的第一人工决策信息作为目标决策信息, 所述 目标决策信息用于对所述车辆进行控制。 [0012] 本公开第二方面提供一种智能驾驶装置, 包括:
[0013] 第一确定模块, 用于确定预设的人工智能感知模型的第一置信度, 所述人工智 能感知模型用于根据传感器信息输出路况信息, 所述第一置信度表示所述人工 智能感知模型输出正确的路况信息的概率;
[0014] 第一获取模块, 用于当所述第一置信度大于或等于第一预设阈值时, 获取所述 人工智能感知模型根据所述传感器信息输出的路况信息作为目标路况信息; [0015] 第二确定模块, 用于确定预设的人工智能决策模型对所述目标路况信息的第二 置信度, 所述人工智能决策模型用于根据所述目标路况信息输出决策信息, 所 述第二置信度表示所述人工智能决策模型输出正确的决策信息的概率;
[0016] 第一显示模块, 用于若所述第二置信度小于第二预设阈值, 则显示所述目标路 况信息;
[0017] 第二获取模块, 用于将接收到的对所述目标路况信息的第一人工决策信息作为 目标决策信息, 所述目标决策信息用于对所述车辆进行控制。
[0018] 本公开第三方面提供一种计算机可读存储介质, 其上存储有计算机程序指令, 该程序指令被处理器执行时实现本公开第一方面所述方法的步骤。
[0019] 本公开第四方面提供一种智能驾驶装置, 包括: 本公开第三方面所述的计算机 可读存储介质; 以及一个或者多个处理器, 用于执行所述计算机可读存储介质 中的程序。
[0020] 采用上述技术方案, 至少可以达到如下技术效果:
[0021] 基于预设的人工智能感知模型的置信度和预设的人工智能决策模型的置信度来 确定是否触发人工介入决策, 可以使当通过人工智能难以做出可靠决策时, 自 适应地触发人工介入决策来确保得到可靠的决策结果, 基于可靠的决策结果对 车辆进行控制, 可以提升车辆行驶的安全性。
[0022] 本公开的其他特征和优点将在随后的具体实施方式部分予以详细说明。
发明概述
对附图的简要说明
附图说明
[0023] 图 1是根据本公开一示例性实施例示出的一种智能驾驶方法的流程图; [0024] 图 2是根据本公开另一示例性实施例示出的一种智能驾驶方法的流程图;
[0025] 图 3是根据本公开一示例性实施例示出的一种智能驾驶装置的框图;
[0026] 图 4是根据本公开另一示例性实施例示出的一种智能驾驶装置的框图;
[0027] 图 5是根据本公开一示例性实施例示出的一种智能驾驶系统的框图;
[0028] 图 6是根据本公开另一示例性实施例示出的一种智能驾驶装置的框图。
[0029] 具体实施方式
[0030] 为使本公开实施例的目的、 技术方案和优点更加清楚, 下面将结合本公开实施 例中的附图, 对本公开实施例中的技术方案进行清楚、 完整地描述, 显然, 所 描述的实施例是本公开一部分实施例, 而不是全部的实施例。 基于本公开中的 实施例, 本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他 实施例, 都属于本公开保护的范围。
[0031] 需要说明的是, 本公开的说明书和权利要求书以及上述附图中的术语“第一”、
“第二”等是用于区别类似的对象, 而不必理解为描述特定的顺序或先后次序。
[0032] 图 1是根据本公开一示例性实施例示出的一种智能驾驶方法的流程图, 如图 1所 示, 该方法包括以下步骤:
[0033] 在步骤 S11中, 确定预设的人工智能感知模型的第一置信度, 人工智能感知模 型用于根据传感器信息输出路况信息。
[0034] 在本公开的实施例中, 第一置信度表示人工智能感知模型输出正确的路况信息 的概率, 其中, 第一置信度越高, 则表明人工智能感知模型输出正确的路况信 息的概率越大, 因而可认为人工智能感知模型输出的路况信息的可靠性越高。
[0035] 值得说明的是, 对于第一置信度的评价, 可以采用不同的评价方法, 例如相似 度、 分类概率等。
[0036] 此外, 人工智能感知模型可以是卷积神经网络 (Convolutional Neural Network , CNN) 模型。
[0037] 在一种实施方式中, 传感器信息可以通过设置在车辆上适当位置的采集装置采 集, 其中, 采集装置可以包括但不限于: 雷达、 毫米波传感器、 超声波传感器 、 GPS定位器、 图像采集器等等。 相应地, 传感器信息可以包括但不限于声音、 距离、 图像等信息。 [0038] 在步骤 S12中, 当第一置信度大于或等于第一预设阈值时, 获取人工智能感知 模型根据传感器信息输出的路况信息作为目标路况信息。
[0039] 当第一置信度大于或等于第一预设阈值时, 可以认为人工智能感知模型根据传 感器信息输出的路况信息的可靠性较高, 因此可以触发人工智能进行感知, 即 获取人工智能感知模型根据传感器信息输出的路况信息作为目标路况信息。
[0040] 在本公开的实施例中, 路况信息可以包括但不限于车辆周围的行人、 其余车辆 、 车道线以及交通标识等信息。
[0041] 在步骤 S13中, 确定预设的人工智能决策模型的第二置信度, 人工智能决策模 型用于根据目标路况信息输出决策信息。
[0042] 在本公开的实施例中, 第二置信度表示人工智能决策模型根据目标路况信息输 出正确的决策信息的概率, 其中, 第二置信度越高, 表明人工智能决策模型输 出正确的决策信息的概率越大, 因而可以认为人工智能决策模型输出的决策信 息的可靠性越高。
[0043] 值得说明的是, 对于第二置信度的评价, 可以采用不同的评价方法, 例如相似 度、 分类概率等。
[0044] 此外, 人工智能决策模型可以是循环神经网络 (Recurrent Neural
Network, RNN) 模型。
[0045] 在步骤 S14中, 若第二置信度小于第二预设阈值, 则显示目标路况信息。
[0046] 在步骤 S15中, 将接收到的对目标路况信息的第一人工决策信息作为目标决策 信息, 目标决策信息用于对车辆进行控制。
[0047] 若第二置信度小于第二预设阈值, 则可以认为人工智能决策模型根据目标路况 信息输出的决策信息的可靠性较低, 为了保证决策信息的可靠性, 此时可以触 发人工介入决策, 即可以向目标用户展示目标路况信息, 由目标用户根据目标 路况信息进行决策并输入相应的决策信息 (即第一人工决策信息) 。 在接收到 对目标路况信息的第一人工决策信息后, 可以将第一人工决策信息作为目标决 策信息, 并根据该目标决策信息来对车辆进行控制, 例如控制车辆加速、 减速 以及转弯等等。 其中, 第一决策信息可以包括但不限于车辆的加速度、 速度、 行驶方向等等, 目标用户可以包括但不限于驾乘者、 远程监控车辆的客服人员 、 操控人员等等。
[0048] 在一种实施方式中, 可以采用构建抽象化图像的方式对目标路况信息进行显示 。 示例地, 可以根据目标路况信息确定路况参数 (例如包括但不限于车辆周围 的障碍物与车辆之间的相对位置关系以及车辆前方的车道线、 交通标识等) , 并将路况参数发送给目的端 (例如, 车载显示设备、 远程监控设备、 目标用户 携带的手机以及佩戴的头盔等) , 由目的端根据路况参数重建路况图像并向目 标用户展示该路况图像, 以便目标用户根据该路况图像进行决策。 这样, 相对 于向目的端直接传输路况图像的方式, 可以减少带宽, 提高传输速率。
[0049] 采用上述智能驾驶方法, 基于预设的人工智能感知模型的置信度和预设的人工 智能决策模型的置信度来确定是否触发人工介入决策, 可以使在通过人工智能 难以做出可靠决策时, 自适应地触发人工介入决策来确保得到可靠的决策结果 , 基于可靠的决策结果对车辆进行控制, 可以提升车辆行驶的安全性。
[0050] 在另一个实施例中, 如图 2所示, 在接收到对目标路况信息的第一人工决策信 息后, 还可以利用目标路况信息和第一人工决策信息更新人工智能决策模型, 以优化和提升该人工智能决策模型的性能, 以便在下一次得到相同或相类似的 目标路况信息时, 提升人工智能决策模型根据目标路况信息输出的决策信息的 可靠性, 进一步减少人工介入, 达到节省人力的目的。 具体地, 可以将目标路 况信息和第一人工决策信息作为数据样本对来训练人工智能决策模型。
[0051] 在另一个实施例中, 如图 2所示, 若第二置信度大于或等于第二预设阈值, 则 可以认为人工智能决策模型输出的决策信息的可靠性较高, 此时可以触发人工 智能决策, 即获取人工智能决策模型根据目标路况信息输出的决策信息作为目 标决策信息, 并利用该目标决策信息对车辆进行控制。
[0052] 在另一个实施例中, 如图 2所示, 当第一置信度小于第一预设阈值时, 可以认 为人工智能感知模型输出的路况信息的可靠性较低, 此时可以触发人工介入感 知和决策, 即可以向目标用户展示传感器信息, 由目标用户根据采集到的传感 器信息判断路况并基于判断结果进行决策, 且输入判断结果 (人工感知信息) 和决策信息 (即第二人工决策信息) 。 在接收到对传感器信息的人工感知信息 和第二人工决策信息后, 可以将第二人工决策信息作为目标决策信息, 根据该 目标决策信息对车辆进行控制, 例如, 控制车辆加速、 减速、 转弯等等。 其中 , 第二决策信息可以包括但不限于车辆的速度、 加速度、 行驶方向等等, 目标 用户可以包括但不限于驾乘者、 远程监控车辆的客服人员、 操控人员等等。
[0053] 在一种实施方式中, 为了便于人工感知和决策, 显示传感器信息的方式可以包 括增强现实 (Augmented Reality, AR) 、 虚拟现实 (Virtual Reality, VR) 以及 混合现实 (Mixed Reality, MR) 等方式, 以辅助目标用户感知和决策。
[0054] 在另一个实施例中, 如图 2所示, 为了进一步提升人工智能感知模型和人工智 能决策模型的性能, 还可以利用传感器信息和接收到的人工感知信息更新人工 智能感知模型, 以及利用人工感知信息和第二人工决策信息更新人工智能决策 模型。 具体地, 对于人工智能感知模型, 可以将传感器信息和人工感知信息标 注为数据样本对来训练人工智能感知模型; 对于人工智能决策模型, 可以将人 工感知信息和第二人工决策信息作为数据样本对来训练人工智能决策模型。
[0055] 采用上述方法, 基于人工智能感知模型的第一置信度和人工智能决策模型的第 二置信度, 在第一置信度小于第一预设阈值时, 触发人工介入感知和决策; 而 在第一置信度大于或等于第一预设阈值而第二置信度小于第二预设阈值时, 触 发人工介入决策, 通过自适应地引入两级人工介入机制来确保得到可靠的决策 结果, 进而解决了目前仅依靠人工智能难以确保车辆行驶安全性的问题。 并且 , 通过人工介入数据来更新人工智能感知模型和人工智能决策模型, 可以优化 和提升人工智能的性能, 以提高人工智能感知结果和人工智能决策结果的可靠 性, 在进一步确保车辆行驶安全性的同时, 可以进一步减少人工介入, 达到节 省人力的目的。
[0056] 值得说明的是, 本公开上述实施例所示的智能驾驶方法可以优选地应用于云端 服务器。
[0057] 图 3是根据本公开一示例性实施例示出的一种智能驾驶装置的框图, 如图 3所示 , 该装置 300可以包括:
[0058] 第一确定模块 301, 用于确定预设的人工智能感知模型的第一置信度, 所述人 工智能感知模型用于根据传感器信息输出路况信息, 所述第一置信度表示所述 人工智能感知模型输出正确的路况信息的概率; [0059] 第一获取模块 302, 用于当所述第一置信度大于或等于第一预设阈值时, 获取 所述人工智能感知模型根据所述传感器信息输出的路况信息作为目标路况信息
[0060] 第二确定模块 303 用于确定预设的人工智能决策模型的第二置信度, 所述人 工智能决策模型用于根据所述目标路况信息输出决策信息, 所述第二置信度表 示所述人工智能决策模型输出正确的决策信息的概率;
[0061] 第一显示模块 304, 用于若所述第二置信度小于第二预设阈值, 则显示所述目 标路况信息;
[0062] 第二获取模块 305 用于将接收到的对所述目标路况信息的第一人工决策信息 作为目标决策信息, 所述目标决策信息用于对车辆进行控制。
[0063] 在另一个实施例中, 如图 4所示, 该装置 300还包括:
[0064] 第一更新模块 306 , 用于利用所述目标路况信息和所述第一人工决策信息更新 所述人工智能决策模型。
[0065] 在另一个实施例中, 如图 4所示, 该装置 300还包括:
[0066] 第三获取模块 307 , 用于若所述第二置信度大于或等于所述第二预设阈值, 则 获取所述人工智能决策模型根据所述目标路况信息输出的决策信息作为目标决 策信息。
[0067] 在另一个实施例中, 如图 4所示, 该装置 300还包括:
[0068] 第二显示模块 308 , 用于当所述第一置信度小于所述第一预设阈值时, 显示所 述传感器信息;
[0069] 第四获取模块 309 , 用于接收对所述传感器信息的人工感知信息和第二人工决 策信息并将所述第二人工决策信息作为目标决策信息。
[0070] 在另一个实施例中, 如图 4所示, 该装置 300还包括:
[0071] 第二更新模块 310, 用于利用所述传感器信息和所述人工感知信息更新所述人 工智能感知模型;
[0072] 第三更新模块 311 用于利用所述人工感知信息和所述第二人工决策信息更新 所述人工智能决策模型。
[0073] 在另一个实施例中, 如图 4所示, 所述第一显示模块 304包括: [0074] 第一确定子模块 341 用于根据所述目标路况信息确定路况参数;
[0075] 第一显示子模块 342, 用于将所述路况参数发送给目的端, 以指示所述目的端 根据所述路况参数生成并显示路况图像。
[0076] 在另一个实施例中, 所述显示所述传感器信息的方式包括: 增强现实、 虚拟现 实和混合现实。
[0077] 关于上述实施例中的装置, 其中各个模块执行操作的具体方式已经在有关该方 法的实施例中进行了详细描述, 此处将不做详细阐述说明。
[0078] 采用上述智能驾驶装置, 基于人工智能感知模型的第一置信度和人工智能决策 模型的第二置信度, 在第一置信度小于第一预设阈值时, 触发人工介入感知和 决策; 而在第一置信度大于或等于第一预设阈值而第二置信度小于第二预设阈 值时, 触发人工介入决策, 通过自适应地引入两级人工介入机制来确保得到可 靠的决策结果, 进而解决了目前仅依靠人工智能难以确保车辆行驶安全性的问 题。 并且, 通过人工介入数据来更新人工智能感知模型和人工智能决策模型, 可以优化和提升人工智能的性能, 以提高人工智能感知结果和人工智能决策结 果的可靠性, 在进一步确保车辆行驶安全性的同时, 可以进一步减少人工介入 , 达到节省人力的目的。
[0079] 本公开实施例还提供了一种车辆, 所述车辆可以包括信息采集装置和本公开上 述实施例提供的智能驾驶装置。
[0080] 本公开实施例还提供了一种智能驾驶系统, 参见图 5 , 该智能驾驶系统可以包 括: 云端服务器 500、 显示设备 600以及位于现场的车辆控制装置 700。
[0081] 云端服务器 500包括智能驾驶装置 510。 显示设备 600可以例如是显示屏、 虚拟 现实头盔、 增强现实头盔等等。 车辆控制装置 700可以为车辆、 手机、 电脑等终 端, 其可以包括信息采集装置 710和决策执行装置 720, 其中, 信息采集装置 710 可以例如是图像采集器、 超声波传感器、 毫米波传感器、 GPS定位器、 雷达等等 , 其可用于采集车辆周围环境的传感器信息。 其中, 传感器信息可以例如包括 但不限于声音、 图像、 距离等信息。
[0082] 在车辆的行驶过程中, 信息采集装置 710可以通过有线 (例如, 电缆、 电网) 或者无线 (例如, 蓝牙、 WIFI) 的方式将采集到的传感器信息发送给云端服务 器 500, 由云端服务器 500的智能驾驶装置 510对传感器信息进行分析得到路况信 息并根据路况信息进行决策, 将决策信息发送给车辆控制装置 700的决策执行装 置 720, 由决策执行装置 720根据决策信息对车辆进行控制。 其中, 路况信息可 以包括但不限于: 车辆周围的行人、 其余车辆、 车道线、 交通标识等信息。 决 策信息可以例如包括控制车辆加速、 减速、 转弯等信息。
[0083] 其中, 参见图 6 , 该智能驾驶装置 510可以包括处理器 522, 其数量可以为一个 或多个, 以及存储器 532, 用于存储可由处理器 522执行的计算机程序。 存储器 5 32中存储的计算机程序可以包括一个或一个以上的每一个对应于一组指令的模 块。 此外, 处理器 522可以被配置为执行该计算机程序, 以执行上述的智能驾驶 方法。
[0084] 另外, 该智能驾驶装置 510还可以包括电源组件 526和通信组件 550, 该电源组 件 526可以被配置为执行智能驾驶装置 510的电源管理, 该通信组件 550可以被配 置为实现智能驾驶装置 510的通信, 例如, 有线或无线通信。 此外, 该智能驾驶 装置 510还可以包括输入 /输出 (I/O) 接口 558。 智能驾驶装置 510可以操作基于 存储在存储器 532的操作系统, 例如 Windows ServerTM, Mac OS
XTM, UnixTM, LinuxTM等等。
[0085] 在另一示例性实施例中, 还提供了一种包括程序指令的计算机可读存储介质, 该程序指令被处理器执行时实现上述的智能驾驶方法的步骤。 例如, 该计算机 可读存储介质可以为上述包括程序指令的存储器 532, 上述程序指令可由智能驾 驶装置 510的处理器 522执行以完成上述的智能驾驶方法。
[0086] 以上结合附图详细描述了本公开的优选实施方式, 但是, 本公开并不限于上述 实施方式中的具体细节, 在本公开的技术构思范围内, 可以对本公开的技术方 案进行多种简单变型, 这些简单变型均属于本公开的保护范围。
[0087] 另外需要说明的是, 在上述具体实施方式中所描述的各个具体技术特征, 在不 矛盾的情况下, 可以通过任何合适的方式进行组合, 为了避免不必要的重复, 本公开对各种可能的组合方式不再另行说明。
[0088] 此外, 本公开的各种不同的实施方式之间也可以进行任意组合, 只要其不违背 本公开的思想, 其同样应当视为本公开所公开的内容。

Claims

权利要求书
[权利要求 1] 一种智能驾驶方法, 其特征在于, 包括:
确定预设的人工智能感知模型的第一置信度, 所述人工智能感知模型 用于根据传感器信息输出路况信息, 所述第一置信度表示所述人工智 能感知模型输出正确的路况信息的概率;
当所述第一置信度大于或等于第一预设阈值时, 获取所述人工智能感 知模型根据所述传感器信息输出的路况信息作为目标路况信息; 确定预设的人工智能决策模型的第二置信度, 所述人工智能决策模型 用于根据所述目标路况信息输出决策信息, 所述第二置信度表示所述 人工智能决策模型输出正确的决策信息的概率; 若所述第二置信度小于第二预设阈值, 则显示所述目标路况信息; 将接收到的对所述目标路况信息的第一人工决策信息作为目标决策信 息, 所述目标决策信息用于对车辆进行控制。
[权利要求 2] 根据权利要求 1所述的方法, 其特征在于, 所述方法还包括:
利用所述目标路况信息和所述第一人工决策信息更新所述人工智能决 策模型。
[权利要求 3] 根据权利要求 1所述的方法, 其特征在于, 所述方法还包括:
若所述第二置信度大于或等于所述第二预设阈值, 则获取所述人工智 能决策模型根据所述目标路况信息输出的决策信息作为目标决策信息
[权利要求 4] 根据权利要求 1所述的方法, 其特征在于, 所述方法还包括;
当所述第一置信度小于所述第一预设阈值时, 显示所述传感器信息; 接收对所述传感器信息的人工感知信息和第二人工决策信息并将所述 第二人工决策信息作为目标决策信息。
[权利要求 5] 根据权利要求 4所述的方法, 其特征在于, 所述方法还包括:
利用所述传感器信息和所述人工感知信息更新所述人工智能感知模型 利用所述人工感知信息和所述第二人工决策信息更新所述人工智能决 策模型。
[权利要求 6] 根据权利要求 1所述的方法, 其特征在于, 所述显示所述目标路况信 息, 包括:
根据所述目标路况信息确定路况参数;
将所述路况参数发送给目的端, 以指示所述目的端根据所述路况参数 生成并显示路况图像。
[权利要求 7] 根据权利要求 4所述的方法, 其特征在于, 所述显示所述传感器信息 的方式包括: 增强现实、 虚拟现实和混合现实。
[权利要求 8] 一种智能驾驶装置, 其特征在于, 包括:
第一确定模块, 用于确定预设的人工智能感知模型的第一置信度, 所 述人工智能感知模型用于根据传感器信息输出路况信息, 所述第一置 信度表示所述人工智能感知模型输出正确的路况信息的概率; 第一获取模块, 用于当所述第一置信度大于或等于第一预设阈值时, 获取所述人工智能感知模型根据所述传感器信息输出的路况信息作为 目标路况信息;
第二确定模块, 用于确定预设的人工智能决策模型的第二置信度, 所 述人工智能决策模型用于根据所述目标路况信息输出决策信息, 所述 第二置信度表示所述人工智能决策模型输出正确的决策信息的概率; 第一显示模块, 用于若所述第二置信度小于第二预设阈值, 则显示所 述目标路况信息;
第二获取模块, 用于将接收到的对所述目标路况信息的第一人工决策 信息作为目标决策信息, 所述目标决策信息用于对车辆进行控制。
[权利要求 9] 根据权利要求 8所述的装置, 其特征在于, 所述装置还包括:
第一更新模块, 用于利用所述目标路况信息和所述第一人工决策信息 更新所述人工智能决策模型。
[权利要求 10] 根据权利要求 8所述的装置, 其特征在于, 所述装置还包括:
第三获取模块, 用于若所述第二置信度大于或等于所述第二预设阈值 , 则获取所述人工智能决策模型根据所述目标路况信息输出的决策信 息作为目标决策信息。
[权利要求 11] 根据权利要求 8所述的装置, 其特征在于, 所述装置还包括:
第二显示模块, 用于当所述第一置信度小于所述第一预设阈值时, 显 示所述传感器信息;
第四获取模块, 用于接收对所述传感器信息的人工感知信息和第二人 工决策信息并将所述第二人工决策信息作为目标决策信息。
[权利要求 12] 根据权利要求 11所述的装置, 其特征在于, 所述装置还包括:
第二更新模块, 用于利用所述传感器信息和所述人工感知信息更新所 述人工智能感知模型;
第三更新模块, 用于利用所述人工感知信息和所述第二人工决策信息 更新所述人工智能决策模型。
[权利要求 13] 根据权利要求 8所述的装置, 其特征在于, 所述第一显示模块包括: 第一确定子模块, 用于根据所述目标路况信息确定路况参数; 第一显示子模块, 用于将所述路况参数发送给目的端, 以指示所述目 的端根据所述路况参数生成并显示路况图像。
[权利要求 14] 根据权利要求 11所述的装置, 其特征在于, 所述显示所述传感器信息 的方式包括: 增强现实、 虚拟现实和混合现实。
[权利要求 15] 计算机可读存储介质, 其上存储有计算机程序, 其特征在于, 该程序 被处理器执行时实现权利要求 1~7中任一项所述方法的步骤。
[权利要求 16] 一种智能驾驶装置, 其特征在于, 包括:
权利要求 15中所述的计算机可读存储介质; 以及 一个或者多个处理器, 用于执行所述计算机可读存储介质中的程序。
PCT/CN2018/101568 2018-08-21 2018-08-21 智能驾驶方法、装置及存储介质 WO2020037500A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN201880001564.3A CN109196437B (zh) 2018-08-21 2018-08-21 智能驾驶方法、装置及存储介质
PCT/CN2018/101568 WO2020037500A1 (zh) 2018-08-21 2018-08-21 智能驾驶方法、装置及存储介质

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2018/101568 WO2020037500A1 (zh) 2018-08-21 2018-08-21 智能驾驶方法、装置及存储介质

Publications (1)

Publication Number Publication Date
WO2020037500A1 true WO2020037500A1 (zh) 2020-02-27

Family

ID=64938508

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/101568 WO2020037500A1 (zh) 2018-08-21 2018-08-21 智能驾驶方法、装置及存储介质

Country Status (2)

Country Link
CN (1) CN109196437B (zh)
WO (1) WO2020037500A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN118035939A (zh) * 2024-02-26 2024-05-14 安徽蔚来智驾科技有限公司 感知目标的置信度获取方法及自动驾驶规划控制方法

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109901574B (zh) * 2019-01-28 2021-08-13 华为技术有限公司 自动驾驶方法及装置
CN110347043B (zh) * 2019-07-15 2023-03-10 武汉天喻信息产业股份有限公司 一种智能驾驶控制方法及装置
CN112857381A (zh) * 2019-11-28 2021-05-28 北京搜狗科技发展有限公司 一种路径推荐方法、装置及可读介质
CN113844463B (zh) * 2021-09-26 2023-06-13 国汽智控(北京)科技有限公司 基于自动驾驶系统的车辆控制方法、装置及车辆
CN116580559A (zh) * 2023-05-18 2023-08-11 中国公路工程咨询集团有限公司 一种基于人工智能技术的交通数据处理方法

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105936294A (zh) * 2015-03-03 2016-09-14 富士重工业株式会社 车辆的行驶控制装置
CN107223261A (zh) * 2016-12-07 2017-09-29 深圳前海达闼云端智能科技有限公司 人机混合决策方法和装置
CN107640159A (zh) * 2017-08-04 2018-01-30 吉利汽车研究院(宁波)有限公司 一种自动驾驶人机交互系统及方法
CN107850895A (zh) * 2015-05-13 2018-03-27 优步技术公司 通过引导协助操作的自动驾驶车辆

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105936294A (zh) * 2015-03-03 2016-09-14 富士重工业株式会社 车辆的行驶控制装置
CN107850895A (zh) * 2015-05-13 2018-03-27 优步技术公司 通过引导协助操作的自动驾驶车辆
CN107223261A (zh) * 2016-12-07 2017-09-29 深圳前海达闼云端智能科技有限公司 人机混合决策方法和装置
CN107640159A (zh) * 2017-08-04 2018-01-30 吉利汽车研究院(宁波)有限公司 一种自动驾驶人机交互系统及方法

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN118035939A (zh) * 2024-02-26 2024-05-14 安徽蔚来智驾科技有限公司 感知目标的置信度获取方法及自动驾驶规划控制方法

Also Published As

Publication number Publication date
CN109196437A (zh) 2019-01-11
CN109196437B (zh) 2021-04-09

Similar Documents

Publication Publication Date Title
WO2020037500A1 (zh) 智能驾驶方法、装置及存储介质
CN110249609B (zh) 用于自主车辆的带宽约束图像处理
US20190329790A1 (en) Systems and Methods for Using Machine Learning to Determine Passenger Ride Experience
US10809719B2 (en) Systems and methods of controlling an autonomous vehicle using an enhanced trajectory following configuration
CN113762252B (zh) 无人机智能跟随目标确定方法、无人机和遥控器
JP7424086B2 (ja) 車両マイクロクラウドによる異常マッピング
WO2020042859A1 (zh) 智能驾驶控制方法和装置、车辆、电子设备、存储介质
US10791436B2 (en) Systems and methods for a vehicle application programming interface
US20210191398A1 (en) System and Methods for Automated Detection of Vehicle Cabin Events for Triggering Remote Operator Assistance
CN106515724B (zh) 一种隧道内行车控制方法及终端
US11242050B2 (en) Reinforcement learning with scene decomposition for navigating complex environments
JP2020502654A (ja) ヒューマンマシンハイブリッド意思決定方法および装置
CN109472884A (zh) 无人车数据存储方法、装置、设备及存储介质
CN111357021A (zh) 用于将自主车辆与乘车者匹配的系统和方法
CN114758502B (zh) 双车联合轨迹预测方法及装置、电子设备和自动驾驶车辆
CN113552867B (zh) 一种运动轨迹的规划方法及轮式移动设备
US20220224860A1 (en) Method for Presenting Face In Video Call, Video Call Apparatus, and Vehicle
CN110091877A (zh) 用于车辆安全驾驶的控制方法、系统及车辆
CN117519206B (zh) 基于生成式扩散模型的自动驾驶模型、方法、装置和车辆
US11691634B1 (en) On-vehicle driving behavior modelling
CN109741601A (zh) 基于区块链的拥堵治理方法、装置、设备和存储介质
CN110456798A (zh) 用于控制车辆行驶的方法及装置
CN116264038A (zh) 信号灯控制方法及装置、电子设备和存储介质
WO2023202313A1 (zh) 位置预测方法、装置、电子设备及存储介质
CN109215365B (zh) 路况监测的方法、装置、设备及存储介质

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: 18930915

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18930915

Country of ref document: EP

Kind code of ref document: A1