WO2022183329A1 - 一种智能驾驶方法、装置、存储介质及计算机程序 - Google Patents

一种智能驾驶方法、装置、存储介质及计算机程序 Download PDF

Info

Publication number
WO2022183329A1
WO2022183329A1 PCT/CN2021/078497 CN2021078497W WO2022183329A1 WO 2022183329 A1 WO2022183329 A1 WO 2022183329A1 CN 2021078497 W CN2021078497 W CN 2021078497W WO 2022183329 A1 WO2022183329 A1 WO 2022183329A1
Authority
WO
WIPO (PCT)
Prior art keywords
vehicle
lane
control
current moment
point
Prior art date
Application number
PCT/CN2021/078497
Other languages
English (en)
French (fr)
Inventor
王旭
王晓祺
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to PCT/CN2021/078497 priority Critical patent/WO2022183329A1/zh
Priority to EP21928420.5A priority patent/EP4296133A4/en
Priority to CN202180000458.5A priority patent/CN113165670A/zh
Publication of WO2022183329A1 publication Critical patent/WO2022183329A1/zh

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W60/00—Drive control systems specially adapted for autonomous road vehicles
    • B60W60/001—Planning or execution of driving tasks
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
    • B60W30/08—Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
    • B60W30/09—Taking automatic action to avoid collision, e.g. braking and steering
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
    • B60W30/18—Propelling the vehicle
    • B60W30/18009—Propelling the vehicle related to particular drive situations
    • B60W30/18163—Lane change; Overtaking manoeuvres
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
    • B60W40/02—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to ambient conditions
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W60/00—Drive control systems specially adapted for autonomous road vehicles
    • B60W60/001—Planning or execution of driving tasks
    • B60W60/0011—Planning or execution of driving tasks involving control alternatives for a single driving scenario, e.g. planning several paths to avoid obstacles
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
    • B60W2050/0001—Details of the control system
    • B60W2050/0043—Signal treatments, identification of variables or parameters, parameter estimation or state estimation
    • B60W2050/0052—Filtering, filters
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W2552/00—Input parameters relating to infrastructure
    • B60W2552/53—Road markings, e.g. lane marker or crosswalk
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W2554/00—Input parameters relating to objects
    • B60W2554/20—Static objects
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W2554/00—Input parameters relating to objects
    • B60W2554/40—Dynamic objects, e.g. animals, windblown objects
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W2554/00—Input parameters relating to objects
    • B60W2554/40—Dynamic objects, e.g. animals, windblown objects
    • B60W2554/404—Characteristics
    • B60W2554/4041—Position
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W2554/00—Input parameters relating to objects
    • B60W2554/40—Dynamic objects, e.g. animals, windblown objects
    • B60W2554/404—Characteristics
    • B60W2554/4042—Longitudinal speed
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W2554/00—Input parameters relating to objects
    • B60W2554/40—Dynamic objects, e.g. animals, windblown objects
    • B60W2554/404—Characteristics
    • B60W2554/4043—Lateral speed

Definitions

  • the present application relates to the technical field of automatic driving, and in particular, to an intelligent driving method, device, storage medium and computer program.
  • the environmental information around the vehicle is obtained through perception and fusion of the sensors of the vehicle, and then the path planning and vehicle control are carried out according to the environmental information, so as to assist the driver in driving.
  • ADAS advanced driver assistance system
  • ADAS Adaptive Cruise Control
  • LKA Lane Keep Assist
  • TJA Traffic Jam Assist
  • ALC Auto Lane Change
  • SLC Semi-Auto Lane Change
  • CAA Collision Avoid Assist
  • LCC Lane Center Control
  • OBF Object follow
  • the control trajectory corresponding to each characteristic is generated based on multiple trajectory lines.
  • the present application proposes an intelligent driving method, device, storage medium and computer program.
  • an embodiment of the present application provides an intelligent driving method, the method includes: acquiring environmental information around the self-vehicle at the current moment and status information of the self-vehicle at the current moment; according to the environmental information and the status information , determine the control point at the current moment; according to the control point, the control trajectory at the historical moment and the state information, generate the control trajectory at the current moment, where the control trajectory represents the trajectory used to guide the vehicle to travel.
  • the control point at the current moment can be determined according to the environmental information around the vehicle at the current moment and the status information of the vehicle at the current moment; and according to the control point, the control trajectory at the historical moment and the status information of the vehicle at the current moment , to generate the control trajectory at the current moment; in this way, the influence of the environmental information on the characteristics is concentrated to the point in the space (that is, the control point), and the control trajectory of the current moment can be generated through the control point to meet the control requirements in different scenarios, And there is no need to process the environmental information in advance to generate multiple trajectories, which greatly reduces the requirements for computing power, thereby improving the performance of the ADAS system.
  • the determining the control point at the current moment according to the environmental information and the state information includes: according to each of the environmental information The preset priority information of the first object and the state information sequentially determine the control point at the current moment.
  • the control point at the current moment is sequentially determined.
  • the priority of the information source realizes the switching between different features, ensuring a smooth transition between multiple features or scenarios, and does not need to use a state machine to switch between different features, reducing the logic complexity.
  • the first object at least includes: a lane line of the own lane, a target vehicle, and a vehicle in an adjacent lane one; wherein, the priority of the lane line of the own lane, the target vehicle, and the vehicle in the adjacent lane are sequentially reduced; the preset priority information of each first object in the environmental information and all The state information is obtained, and the control points at the current moment are sequentially determined, including: in the case that the lane line of the own lane exists, assigning the corresponding preset distance to the lane line of the own lane and meeting the preset conditions The position point is used as the control point at the current moment, and the preset distance is determined based on the state information; if the lane line of the own lane does not exist, in the case that the target vehicle exists, the corresponding , and the position point that meets the preset conditions is used as the control point at the current moment; if the lane line of the own lane
  • the control point at the current time can be determined according to the lane line of the own vehicle's own lane in the environmental information around the own vehicle at the current time, and then the characteristics of LCC and LKA can be realized; CIPV, determines the control point at the current moment, and then realizes the characteristics of TJA, OBF, etc.; can determine the control point at the current moment according to the environmental information around the vehicle at the current moment, the vehicle in the adjacent lane, and then realize the characteristics of TJA, LKA, etc.; in this way, Use the lane line of the own lane, the target vehicle, and the vehicle in the adjacent lane to change the position of the control points in different scenarios according to the priority order, and determine the current moment according to the first object with the highest priority existing in the environmental information.
  • Control points so as to realize multiple characteristics (such as: LCC, LKA, OBF, TJA, etc.), and at the same time, according to the lane line of the own lane, the priority of the target vehicle, and the vehicle in the adjacent lane, to achieve switching between different characteristics, Smooth transitions between multiple features or scenes are guaranteed.
  • characteristics such as: LCC, LKA, OBF, TJA, etc.
  • a third possible implementation manner of the first aspect according to the preset priority of each first object in the environment information information and the state information, and sequentially determine the control point at the current moment, including: obtaining the position point corresponding to the first object with the highest priority; judging whether there is a second object in the preset area, and the second object is the environment The avoidance object of the own vehicle in the information; when the second object exists, offset processing is performed on the position point corresponding to the first object, and the offset position point is used as the control point at the current moment.
  • the sorting method combined with the existing second object, jointly change the position of the control points in different scenes, and determine the control point at the current moment, which can realize multiple characteristics (such as CAA, road edge distance, etc.), and ensure multiple Smooth transitions between features or scenes.
  • the second object includes a hard boundary of the own lane; the preset area is the first
  • the position point corresponding to the object is an area with a preset size in the center; when the second object exists, the position point corresponding to the first object is subjected to offset processing, and the offset position point is used as the current moment
  • the control points of Outside the area of the preset size, and the offset position point is within the own lane; the offset position point is used as the control point at the current moment.
  • the position point corresponding to the first object is subjected to offset processing, so that the hard boundary is outside the area of the preset size centered on the offset position point, And the offset position point is in the own lane, and the offset position point is used as the control point at the current moment. , to avoid colliding with the hard boundary.
  • the position of the control points in different scenes can be changed together, and the control point at the current moment can be determined, which can realize multiple characteristics (such as features such as curb far away), and guarantees a smooth transition between multiple features or scenes.
  • the second object includes an obstacle; the preset area is a freeway in the freeway an area of a preset size in front of the car; when the second object exists, perform offset processing on the position point corresponding to the first object, and use the offset position point as the control point at the current moment, including : When the obstacle exists, perform offset processing on the position point corresponding to the first object according to the weight information corresponding to the obstacle, so that the obstacle is at the position point after the offset is outside the area with the preset size of the center, and the offset position point is in the own lane; wherein, the weight information indicates the degree to which the vehicle avoids the obstacle; the offset position point is taken as the current moment control point.
  • the weight information corresponding to the obstacle is used to indicate the degree to which the self-vehicle avoids the obstacle in different scenarios, and according to the weight information corresponding to the obstacle, Offset the position point corresponding to the first object, so as to realize the position point offset in different scenarios, and use the offset position point as the control point at the current moment.
  • the control points of different scenes can be changed together. The position of the control point at the current moment is determined, thereby realizing multiple features (such as CAA and other features), and ensuring a smooth transition between multiple features or scenes.
  • the value of the weight information is at least determined by the headway THW, the collision time TTC, and the intrusion width IVW A determination of , wherein the value of the weight information is negatively correlated with the THW, negatively correlated with the TTC, and positively correlated with the IVW.
  • the value of weight information is negatively correlated with THW, that is, the smaller the THW, the more obvious the ego vehicle avoids obstacles
  • the value of weight information is negatively correlated with TTC, that is, the smaller the TTC, the more obvious the ego vehicle avoids obstacles.
  • the more obvious the action is the value of the weight information is positively correlated with the IVW, that is, the larger the IVW, the more obvious the action of the ego vehicle to avoid obstacles; in this way, according to the value of the corresponding weight information in different scenarios, change the position of the control point in different scenarios , determines the control point at the current moment, and realizes a smooth transition of different scenes.
  • the generating the control trajectory at the current moment according to the control point, the control trajectory at a historical moment, and the state information includes: according to The control point, the control trajectory at the historical moment, and the state information use a Kalman filter to generate the control trajectory at the current moment.
  • the Kalman filter has temporal continuity, and the position information of the control point at the current moment, the control trajectory at the historical moment and the vehicle state information at the current moment are input into the Kalman filter, and the control point in space is used in the Kalman filter.
  • the continuity in the time domain generates the control trajectory at the current moment, which meets the control requirements in different scenarios, and greatly reduces the computing power requirements and improves the system performance.
  • an embodiment of the present application provides an intelligent driving device, the device includes: an acquisition module for acquiring environmental information around the own vehicle at the current moment and state information of the vehicle at the current moment; a determination module for According to the environmental information and the state information, the control point at the current moment is determined; the generating module is used for generating the control trajectory at the current moment according to the control point, the control trajectory at the historical moment and the status information, and the control The trajectory represents the trajectory used to guide the ego vehicle to travel.
  • the determining module is further configured to: according to preset priority information and the state of each first object in the environment information information, and sequentially determine the control point at the current moment.
  • the first object at least includes a lane line of the own lane, a target vehicle, the neighbor One of the vehicles in the lane; wherein, the priorities of the lane line of the own lane, the target vehicle, and the vehicle in the adjacent lane are decreased in sequence; the determining module is further used for: the lane line of the own lane In the case of existence, the position point corresponding to the preset distance on the lane line of the own lane and meeting the preset condition is used as the control point at the current moment; the preset distance is determined based on the state information; if the If the lane line of the own lane does not exist, if the target vehicle exists, the position point corresponding to the target vehicle and meeting the preset conditions is used as the control point at the current moment; if the lane line of the own lane exists does not exist, and the target vehicle does not exist, then in the case that the vehicle in the adjacent lane exists, the position
  • the determining module is further configured to: acquire the first object with the highest priority The corresponding position point; determine whether there is a second object in the preset area, the second object is the avoidance object of the own vehicle in the environmental information; when the second object exists, the position point corresponding to the first object is Offset processing is performed, and the offset position point is used as the control point at the current moment.
  • the second object includes a hard boundary of the own lane; the preset area is the first The position point corresponding to the object is an area with a preset size in the center; the determining module is further configured to: when the hard boundary of the own lane exists, perform offset processing on the position point corresponding to the first object, so as to make The hard boundary is outside the area of the preset size centered on the offset position point, and the offset position point is within the own lane; the offset position point is used as the control at the current moment point.
  • the second object includes an obstacle; the preset area is a freeway in the freeway an area with a preset size in front of the vehicle; the determining module is further configured to: when the obstacle exists, perform offset processing on the position point corresponding to the first object according to the weight information corresponding to the obstacle, So that the obstacle is outside the area of the preset size centered on the offset position point, and the offset position point is in the own lane; wherein, the weight information indicates that the vehicle avoids The degree of the obstacle; take the offset position as the control point at the current moment.
  • the value of the weight information is at least determined by the headway THW, the collision time TTC, and the intrusion width IVW A determination of , wherein the value of the weight information is negatively correlated with the THW, negatively correlated with the TTC, and positively correlated with the IVW.
  • the generating module is further configured to: according to the control point at the current moment, the control trajectory at the historical moment, and the state information, Using the Kalman filter, the control trajectory at the current moment is generated.
  • embodiments of the present application provide an intelligent driving device, including: a processor; a memory for storing instructions executable by the processor; wherein the processor is configured to implement the above-mentioned first step when executing the instructions
  • One or more intelligent driving methods in one aspect or multiple possible implementation manners of the first aspect are provided.
  • embodiments of the present application provide a non-volatile computer-readable storage medium on which computer program instructions are stored, and when the computer program instructions are executed by a processor, the above-mentioned first aspect or the first aspect is implemented One or several of the multiple possible implementations of the intelligent driving method.
  • embodiments of the present application provide a computer program product including instructions, which, when run on a computer, enables the computer to execute the first aspect or one of multiple possible implementations of the first aspect or several intelligent driving methods.
  • an embodiment of the present application further provides a vehicle, where the vehicle includes the above-mentioned second aspect or one or more of the intelligent driving devices of the various possible implementations of the second aspect.
  • an embodiment of the present application further provides an automatic driving assistance system, the system includes: a preprocessing layer, a planning layer, a decision-making layer, and a control layer; wherein, the preprocessing layer is used to generate the self-vehicle at the current moment the surrounding environment information; the planning layer is used to determine the control point at the current moment according to the environment information and the status information of the vehicle at the current moment; according to the control point, the control trajectory at the historical moment and the status information, The control trajectory at the current moment is generated; the control trajectory represents the trajectory used to guide the driving of the vehicle; the decision-making layer is used to judge whether the control trajectory is suitable for the working state of the vehicle at the current moment; the control The layer is used to generate a control signal at the current moment based on the control trajectory when the control trajectory is applicable to the working state of the vehicle at the current moment, and the control signal is used for assisted driving of the vehicle.
  • the preprocessing layer is used to generate the self-vehicle at the current moment the surrounding environment information
  • the planning layer is further configured to: according to preset priority information of each first object in the environment information and the state information , and sequentially determine the control points at the current moment.
  • the planning layer is further configured to: acquire the position point corresponding to the first object with the highest priority; Determine whether there is a second object in the preset area, the second object is the avoidance object of the own vehicle in the environmental information; when the second object exists, perform offset processing on the position point corresponding to the first object, And use the offset position point as the control point at the current moment.
  • the seventh aspect if there is a new feature, it is only necessary to add a feature for the feature in the preprocessing layer of the system.
  • information such as the priority of adding a specific corresponding object can also be set, and there is no need to newly generate a trajectory line for this characteristic, and there is no need to adjust the state machine, and the system is highly scalable.
  • FIG. 1 shows a schematic diagram of an application scenario to which an intelligent driving method according to an embodiment of the present application is applicable
  • FIG. 2 shows a flowchart of an intelligent driving method according to an embodiment of the present application
  • FIG. 3 shows a flowchart of determining a control point at the current moment according to an embodiment of the present application
  • FIG. 4 shows a schematic diagram of various vehicles invading the own lane according to an embodiment of the present application
  • FIG. 5 shows a schematic diagram of position point offsets in various vehicle intrusion self-lane scenarios in FIG. 4 according to an embodiment of the present application
  • FIG. 6 shows a schematic diagram of a weighted body function according to an embodiment of the present application
  • FIG. 7 shows a schematic diagram of another weighted body function according to an embodiment of the present application.
  • FIG. 8 shows a flowchart of determining a control point at the current moment according to an embodiment of the present application
  • FIG. 9 shows a schematic diagram of implementing OBF characteristics by using control points according to an embodiment of the present application.
  • FIG. 10 shows a schematic diagram of implementing LCC characteristics using control points according to an embodiment of the present application.
  • FIG. 11 shows a schematic diagram of implementing a TJA feature using a control point according to an embodiment of the present application
  • FIG. 12 shows a schematic diagram of implementing ALC/SLC characteristics using control points according to an embodiment of the present application
  • FIG. 13 shows a comparison diagram of intelligent driving according to an embodiment of the present application
  • FIG. 14 shows a schematic diagram of a lateral offset velocity curve according to an embodiment of the present application.
  • FIG. 15 shows a schematic structural diagram of an intelligent driving device according to an embodiment of the present application.
  • FIG. 16 shows an ADAS system architecture diagram according to an embodiment of the present application
  • FIG. 17 shows a schematic structural diagram of another intelligent driving device according to an embodiment of the present application.
  • the present application provides an intelligent driving method, which can be executed by an intelligent driving device, wherein the intelligent driving device can be set independently, can also be integrated in other devices, and can also be implemented by software or a combination of software and hardware.
  • the intelligent driving device can be applied to ADAS or an automated driving system (Automated Driving System, ADS), and can also be applied to device to device communication (D2D), vehicle to communicate with anything (vehicle to everything, V2X), vehicle to vehicle (Vehicle to Vehicle, V2V), long term evolution and vehicle communication (long term evolution-vehicle, LTE-V), long term evolution and machine communication (long term evolution-machine, LTE-M) etc. scene.
  • D2D device to device communication
  • V2X vehicle to everything
  • V2V vehicle to Vehicle
  • V2V Vehicle to Vehicle
  • long term evolution and vehicle communication long term evolution-vehicle, LTE-V
  • long term evolution-machine long term evolution-machine
  • the intelligent driving device may be a vehicle with a control trajectory generation function, or other components with a control trajectory generation function.
  • the intelligent driving device includes but is not limited to: on-board terminal, on-board controller, on-board module, on-board module, on-board components, on-board chip, on-board unit, on-board radar or on-board camera and other sensors, the vehicle can pass the on-board terminal, on-board controller and other sensors.
  • vehicle-mounted module, vehicle-mounted module, vehicle-mounted component, vehicle-mounted chip, vehicle-mounted unit, vehicle-mounted radar or camera to implement the intelligent driving method provided in this application.
  • the intelligent driving device can also be other intelligent terminals with control trajectory generation function other than the vehicle, or be set in other intelligent terminals with control trajectory generation function other than the vehicle, or set in the intelligent terminal. part of the terminal.
  • the intelligent terminal may be other terminal equipment such as intelligent transportation equipment, smart home equipment, robots, and drones.
  • the intelligent driving device includes, but is not limited to, an intelligent terminal or a controller, a chip, a sensor such as a radar or a camera, and other components in the intelligent terminal.
  • the intelligent driving device may be a general-purpose device or a special-purpose device.
  • the intelligent driving device may also be a desktop computer, a portable computer, a network server, a cloud server, a personal digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, an embedded device or other devices with A device that handles functionality.
  • PDA personal digital assistant
  • the intelligent driving device may also be a chip or processor with a processing function, and the intelligent driving device may include multiple processors.
  • the processor can be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor.
  • the chip or processor with processing function may be arranged in the sensor, or may not be arranged in the sensor, but arranged at the receiving end of the output signal of the sensor.
  • the intelligent driving method provided is exemplarily described by taking the application of an intelligent driving device in an ADAS system as an example.
  • FIG. 1 shows a schematic diagram of an application scenario to which an intelligent driving method according to an embodiment of the present application is applicable.
  • the application scenario can be an advanced assisted driving scenario, and the scenario includes: a self-vehicle 401 , at least one sensor (not shown in the figure) is installed in the self-vehicle 401 , and the self-vehicle 401 also includes an intelligent driving device (not shown in the figure), the sensor is used to sense the environmental information around the own vehicle 401, for example, the lane line 403 of the own lane (ie the lane where the own vehicle is located), the adjacent lane (ie the lane adjacent to the own vehicle lane)
  • FIG. 2 shows a flowchart of an intelligent driving method according to an embodiment of the present application.
  • the method can be applied to the intelligent driving device in the self-vehicle 401 in the above-mentioned FIG. 1 , and the method can include the following steps:
  • Step 501 The intelligent driving device acquires the environment information around the own vehicle at the current moment and the status information of the vehicle at the current moment.
  • the environment information represents one or more objects in the surrounding environment of the own vehicle and the attribute information of each object, for example, the lane line of the own lane, the road edge boundary of the road, the lane line of the adjacent lane, the target vehicle, the forward traffic flow, Objects such as side vehicles, rear vehicles, vehicles in adjacent lanes, pedestrians, traffic lights, buildings, trees, construction areas, etc., and attribute information such as position, speed, acceleration, orientation, and distance of each object.
  • the environment information around the self-vehicle can be sensed by sensors such as radar (eg, millimeter-wave radar, lidar, ultrasonic radar), image acquisition device (eg, camera) installed around the body of the self-vehicle.
  • the state information of the ego vehicle may include: driving state information, motion state information, etc., wherein the motion state information represents the information of the kinematic state of the ego vehicle, for example, the speed, acceleration, yaw rate, steering angle, steering angle of the ego vehicle Information such as radius, body position and attitude.
  • the running status information can be acquired in real time through sensors installed inside the vehicle.
  • the driving status information represents parameter information set by the driver, for example, the cruising speed set by the driver, the following distance, the headway, the driver's lane change command, the lane change duration, and the like.
  • the driving and riding status information input by the driver through the display screen, the physical buttons of the steering wheel, and the voice can be received.
  • the speed, acceleration, distance, etc. of each object in the above-mentioned surrounding environment may be the relative speed, relative acceleration, and relative distance of each object relative to the vehicle;
  • the position, orientation, etc. of each object may be the vehicle body coordinate system.
  • the position or orientation in the vehicle body coordinate system is used to describe the relative positional relationship between the objects around the own vehicle and the own vehicle, and the relationship between the coordinate system of each sensor and the vehicle body coordinate system can be clearly defined by pre-calibration.
  • the transformation relationship is used to determine the coordinates of each object in the vehicle body coordinate system according to the data collected by the sensor.
  • the position of each object refers to the coordinates in the vehicle body coordinate system;
  • the forward direction is the positive direction of the X-axis
  • the leftward along the rear axle axis is the positive direction of the Y-axis
  • the upward is the positive direction of the Z-axis, that is, the corresponding vehicle body coordinate system XYZ is formed.
  • Step 502 the intelligent driving device determines the control point at the current moment according to the environmental information around the vehicle and the state information of the vehicle at the current moment.
  • the control point also known as the virtual point, may be a point on the object in the surrounding environment of the vehicle, for example, the midpoint of the rear bumper of the target vehicle, or the lane line of the vehicle lane and the vehicle. Points a certain length away, and so on.
  • the control point may also be a point other than the object in the surrounding environment of the self-vehicle, for example, a point between the left and right lane lines of the self-vehicle, or a point at a certain distance directly in front of the self-vehicle, etc. Wait.
  • the unique control point at the current moment may be determined according to the first object contained in the information about the surrounding environment of the vehicle at the current moment and the status information of the vehicle (eg, driving and riding status information).
  • the control points at the current moment may be sequentially determined according to the preset priority information of each first object in the environmental information and the driving and riding status information at the current moment.
  • the first object with the highest priority existing in the environmental information may be determined according to the preset priority information of each first object, and the control point at the current moment may be determined according to the first object with the highest priority.
  • the preset priority information of each first object it can be determined in turn whether each first object exists in the environmental information around the own vehicle at the current moment; when it is determined that a first object is in the environmental information around the own vehicle at the current moment When it exists, the control point at the current moment is determined according to the existing first object.
  • the first object may include a reference object of the own vehicle, for example, the lane line of the own lane, the target vehicle, the vehicle in the adjacent lane, etc.; different first objects correspond to different characteristics, and each first object corresponds to The characteristics of the can be OBF, LKA, TJA, LCC and other characteristics. That is, the preset priority may also represent the preset priority ordering of processing features in the ADAS system. Exemplarily, the preset priorities of the lane line of the own lane, the target object, and the lane line of the adjacent lane are sequentially lowered. In this way, according to the preset priority information and driving status information of each first object in the environmental information, the control point at the current moment is determined in turn.
  • the priority of the source realizes switching between different features, ensuring a smooth transition between multiple features or scenes. There is no need to use a state machine to switch between different features, which reduces the logic complexity.
  • one or more features may be included, and different features correspond to different first objects, wherein the number of features, the type of the first object, and the preset priority of each type It can be pre-configured by the ADAS system or set by the driver.
  • the first object may include: the lane line of the own lane, the closest vehicle (closest in-path vehicle, CIPV) in the lane where the own vehicle is located (that is, the target vehicle), and the vehicle in the adjacent lane;
  • the priority of the lane line, CIPV, and vehicles in adjacent lanes decreases in turn.
  • the position point corresponding to the preset distance on the lane line of the own lane and meeting the preset conditions is used as the control point at the current moment; If the CIPV exists, the position point corresponding to the CIPV and meeting the preset conditions is used as the control point at the current moment; if the CIPV does not exist, if the vehicle in the adjacent lane exists, the vehicle in the adjacent lane is used as the control point at the current moment.
  • the position point projected on the own lane and meeting the preset conditions is used as the control point at the current moment.
  • the existence of the lane line of the own lane, the existence of CIPV, the existence of vehicles in the adjacent lane, etc. indicating that the sensor of the own vehicle can perceive the data of the lane line of the own lane, CIPV and vehicles in the adjacent lane, etc.
  • the data meets the preset requirements. Requirements for data stability and accuracy.
  • the preset conditions may include: the position point is within the reachable range of the vehicle in the current driving state, for example, the vehicle is currently in a forward driving state, and the preset condition is that the X-axis coordinate value of the position point is greater than 3m, that is, if the position If the point is located in front of the vehicle and the X-axis coordinate value is greater than 3m, the preset conditions are met. If the point is located behind or on the side of the vehicle, the preset conditions are not met.
  • the preset distance can be determined based on the driving and riding status information.
  • the preset distance can include: the X-axis coordinate value of the position point is 15-40m, etc., which is taken from The position point between the lane line of the lane and the longitudinal distance of the vehicle is 15-40m away.
  • the position point corresponding to the preset distance on the lane line of the own lane can be the point on the preset distance on the lane line of the own lane, or the point on the preset distance on the lane line of the own lane can be projected onto the lane center line of the own lane
  • the position point corresponding to the CIPV can be the middle point of the CIPV rear bumper, or the point on the center line of the lane where the CIPV is projected to the own lane, and so on.
  • the position of the projection of the vehicle in the adjacent lane on the own lane can be the projection point of the lateral projection of the rear bumper of the vehicle in the adjacent lane to the own lane.
  • the horizontal direction is the direction perpendicular to the driving direction indicated by the lane line, and correspondingly, the longitudinal direction is the driving direction indicated by the lane line.
  • FIG. 3 shows a flowchart of determining a control point at the current time according to an embodiment of the present application.
  • the intelligent driving device first determines whether there is a lane line of the own lane in the environmental information. When the lane line exists, the position point corresponding to the preset distance on the lane line of the self-lane is obtained. If the position point satisfies the preset conditions, the position point is used as the control point at the current moment. If the conditions are set, it will exit the ADAS system and remind the driver to take over the vehicle control; if the lane line of the own lane does not exist, it will further judge whether there is CIPV in the environmental information.
  • the position point corresponding to the CIPV is obtained, if the position point satisfies the preset conditions, the position point is used as the control point at the current moment, if the position point does not meet the preset conditions, the ADAS system is exited, And remind the driver to take over the vehicle control; if the CIPV does not exist, it will further judge whether there is a vehicle in the adjacent lane. When there is a vehicle in the adjacent lane, obtain the projected position point of the vehicle in the adjacent lane on the own lane. If the position point satisfies the preset conditions, the position point is used as the control point at the current moment.
  • the ADAS system is exited and the driver is reminded to take over vehicle control.
  • the ADAS system is exited and the driver is reminded to take over the vehicle control.
  • the control trajectory is generated, which can realize the characteristics of LCC and LKA.
  • the control trajectory is generated.
  • the control trajectory can be generated, and the characteristics of TJA and LKA can be realized.
  • the positions of the control points in different scenarios can be changed in a prioritized manner, and the current location can be determined according to the first object with the highest priority existing in the environmental information.
  • the control point at the current time can be used to realize multiple characteristics.
  • the control point at the current time can be determined according to the lane line of the own lane in the environmental information around the vehicle at the current time, and then the characteristics of LCC and LKA can be realized.
  • CIPV in the surrounding environmental information determines the control point at the current moment, and then realizes the characteristics of TJA and OBF; it can determine the control point at the current moment according to the vehicles in the adjacent lane of the vehicle in the environmental information around the vehicle at the current moment, and then realize TJA, LKA and other characteristics; at the same time, according to the lane line of the own lane, the priority of the target vehicle, and the vehicle in the adjacent lane, the switching between different characteristics is realized, which ensures a smooth transition between multiple characteristics or scenes.
  • the unique control point at the current moment may be determined according to the first object, the second object and the status information (eg, driving and riding status information) of the vehicle contained in the environmental information around the vehicle at the current moment.
  • the status information eg, driving and riding status information
  • sequentially determining the control points at the current moment may include: obtaining the highest priority The position point corresponding to the first object; determine whether there is a second object in the preset area; when the second object exists, perform offset processing on the position point corresponding to the first object, and use the offset position point as the current moment control point.
  • the position point corresponding to the first object may be the position point corresponding to the preset distance on the lane line of the own lane, the position point corresponding to the CIPV, the position point projected by the vehicle in the adjacent lane on the own lane, etc.
  • the second object is
  • the avoidance objects of the own vehicle in the environmental information can be side vehicles, rear vehicles, pedestrians, construction areas, road borders, etc.
  • Different second objects correspond to different characteristics, and the characteristics corresponding to each second object It can be features such as CAA, curb distance, etc.
  • the preset area is an area of a certain size in the road, for example, it may be a circular area of a certain size centered on the position point corresponding to the first object, or may be a rectangular area of a certain size in the own lane.
  • a control trajectory is generated based on the position point corresponding to the first object obtained above, and the vehicle is controlled to drive according to the control trajectory, there is a risk of colliding with the second object.
  • Offset processing is performed on the position point corresponding to the first object, for example, horizontal offset processing and vertical offset processing, and the offset position point is used as the control point at the current moment, so as to avoid collision with the second object.
  • the position of the control point in different scenes can be changed together, and the control point at the current moment can be determined, which can realize multiple characteristics (such as CAA, road edge away, etc. features), and ensures smooth transitions between multiple features or scenes.
  • one or more features may be included, and different features correspond to different second objects, wherein the number of features, the type of the second object, and the size of the preset area can be determined by The ADAS system is preconfigured or programmed by the driver.
  • the second object includes a hard boundary of a lane;
  • the preset area is a first area with a preset size centered on a position point corresponding to the first object.
  • offset processing is performed on the position point corresponding to the first object, and the offset position point is used as the control point at the current moment, including: When there is a hard boundary of the lane in the area, the position point corresponding to the first object is offset, so that the hard boundary is outside the area of the preset size centered on the offset position point, and the offset position The point is in the own lane; the offset position point is used as the control point at the current moment.
  • the hard boundary may be a boundary that the self-vehicle cannot touch or cross, for example, it may include: a solid line of the self-lane lane, a road edge boundary of a road, etc., and the first area may be centered on the position point corresponding to the first object
  • the area with a certain size may be a circular area formed by taking the position point corresponding to the first object as the center of the circle and the radius of 0.5-1.875 meters.
  • the offset processing can be based on the position of the lane centerline of the own lane, moving the position point corresponding to the first object in the lateral direction by a certain distance.
  • the lane centerline of the own lane is located to the left of the position point corresponding to the first object, then Move the position point corresponding to the first object to the left by a certain distance in the lateral direction, so that the hard boundary is outside the area with a certain size centered on the offset position point, and the offset position point is within In the own lane; the offset position point is used as the control point at the current moment, and the control point is far away from the hard boundary of the lane, so that the vehicle can avoid colliding with the hard boundary when driving according to the control trajectory generated by the control point.
  • the position of the control point in different scenes can be changed together, and the control point at the current moment can be determined. for smooth transitions between multiple features or scenes.
  • the second object may include obstacles such as vehicles on the side, pedestrians, etc.; the preset area may be a second area with a preset size in front of the vehicle in the own lane.
  • offset processing is performed on the position point corresponding to the first object, and the offset position point is used as the control point at the current moment, including: in the second object When there is an obstacle in the area, the position point corresponding to the first object is offset according to the weight information corresponding to the obstacle, so that the obstacle is outside the area with the preset size centered on the offset position point , and the offset position point is in the own lane; the offset position point is used as the control point at the current moment.
  • the second area may be a rectangular area in front of the self-vehicle in the self-vehicle having a certain longitudinal length (for example, 10-40 m) and a certain horizontal width (for example, 1.875-3.75 m).
  • the weight information corresponding to the obstacle indicates the degree to which the self-vehicle avoids the obstacle. For example, the larger the value of the weight information, the more obvious the self-vehicle's action to avoid the obstacle, and the corresponding offset position point corresponds to the first object. The farther the distance between the position points is; the smaller the value of the weight information, the smaller the self-vehicle's action to avoid the obstacle, and the greater the distance between the corresponding offset position point and the position point corresponding to the first object.
  • the weight information corresponding to the obstacles is used to indicate that the ego vehicle avoids the obstacles in different scenarios. and according to the weight information corresponding to the obstacle, the position point corresponding to the first object is offset, so as to realize the position point offset in different scenarios, and the offset position point is used as the control point at the current moment, The control point is far away from the obstacle, so that when the vehicle travels according to the control trajectory generated by the control point, it avoids colliding with the obstacle.
  • the weight information is used to jointly change the position of the control points in different scenes, and determine the control point at the current moment, so as to realize multiple features (such as CAA and other features), and ensure a smooth transition between multiple features or scenes.
  • the value of the weight information may be determined by a preset weight function (or a weight body function), and the parameters included in the weight function may include: headway (Time of Headway, THW), time to collision (Time to Collision) , TTC), invasion width (Invade Vehicle Width, IVW) and so on.
  • THW can be determined from the driving and riding status information, that is, it can be preset by the driver
  • TTC can be determined according to the relative distance and relative speed between the intruding vehicle and the own vehicle in the environmental information
  • IVW can be determined according to the lane line of the own lane in the environmental information. location and the location of the intruding vehicle.
  • ⁇ the value of the weight information
  • f the weight body function
  • the value ⁇ of the information is negatively correlated with THW, that is, the smaller the THW, the more obvious the action of the ego vehicle to avoid obstacles
  • the value ⁇ of the weight information is negatively correlated with the TTC, that is, the smaller the TTC, the more obvious the action of the ego vehicle to avoid obstacles.
  • the value ⁇ of the weight information is positively correlated with the IVW, that is, the larger the IVW, the more obvious the self-vehicle’s action to avoid obstacles; in this way, according to the value of the corresponding weight information in different scenarios, change the position of the control point in different scenarios to determine the current The control points at the moment realize the smooth transition of different scenes.
  • the following takes the obstacle as an intruding vehicle as an example to describe the weight information corresponding to the obstacle in different scenarios.
  • FIG. 4 shows a schematic diagram of various vehicles invading the own lane according to an embodiment of the present application.
  • the direction of the arrow represents the speed direction of the vehicle, wherein the own vehicle 401 and The direction of the arrow in front of the front of the intruding vehicle 701 is the forward direction, and the direction of the arrow on the rear of the intruding vehicle 701 is the reverse direction.
  • the number of arrows indicates the speed of the vehicle.
  • Figure 4(a) shows the scene where the intruding vehicle is kept at a close distance, that is, the speed directions of the intruding vehicle 701 and the own vehicle 401 are both positive and approximately equal;
  • the speed direction of the intruding vehicle 701 and the self-vehicle 401 are both positive, and the speed of the intruding vehicle 701 is greater than that of the self-vehicle 401;
  • the speed direction of the vehicle 701 is reverse, the speed direction of the own vehicle 401 is forward, and the speed of the intruding vehicle 701 is smaller than that of the own vehicle 401;
  • Figure 4(d) shows the scene where the intruding vehicle cuts in at a constant speed, that is, the intrusion
  • the speed directions of the vehicle 701 and the own vehicle 401 are both positive and form a certain angle, and the speed of the intruding vehicle 701 is equal to the speed of the own vehicle 401;
  • the speed directions of the ego vehicle 401 are all positive and form a
  • FIG. 5 shows a schematic diagram of position point offsets in the scenario of various vehicles intruding into the own lane in FIG. 4 according to an embodiment of the present application.
  • the “0” point represents a preset on the lane center line of the own lane
  • the distance point, "1" point represents the lateral projection point of the rear bumper of the intruding vehicle on a lane line away from the own lane of the intruding vehicle.
  • the ego car will not produce evasive action
  • the "1" point is used as the offset position point, that is, "1”
  • the point is used as a control point to generate a control trajectory and guide the ego car to drive, and the ego car will have obvious avoidance action.
  • the self-vehicle will generate an avoidance action, and the degree of avoidance (or the magnitude of the avoidance action) is negatively correlated with the distance of the offset position point relative to 1 point, that is, the closer the offset position point is to 1 point, the greater the avoidance action amplitude. big.
  • the value of the weight information is between 0 and 1, and the difference between "0" and "1" The line segments between them are divided into equal proportions.
  • the value of the weight information is equal to 1, the corresponding "1" point is the offset position; when the value of the weight information is equal to "0", the corresponding "0" point is the offset position.
  • the position point of when the value of the weight information is greater than 0 and less than 1, the point at the corresponding position in the line segment is selected.
  • the offset position point can be determined according to THW, TTC, IVW and the weight body function f. If the offset position point is used as the control point at the current moment to generate the control trajectory at the current moment, It can be ensured that the self-vehicle will not collide with the intruding vehicle along the control trajectory.
  • Fig. 6 shows a schematic diagram of a weight body function according to an embodiment of the present application.
  • FIG. 7 shows a schematic diagram of another weight body function according to an embodiment of the present application.
  • FIG. 8 shows a flowchart of determining a control point at the current time according to an embodiment of the present application.
  • the intelligent driving device first determines whether there is a lane line of the own lane in the environmental information. In the case of existence, the position point corresponding to the preset distance on the lane line of the own lane is obtained; if the lane line of the own lane does not exist, it is further judged whether there is a CIPV in the environmental information. When the CIPV exists, the position point corresponding to the CIPV is obtained; when the CIPV does not exist, it is further judged whether the vehicle exists in the adjacent lane. When there is a vehicle in the adjacent lane, obtain the projected position point of the vehicle in the adjacent lane on the own lane; when the vehicle in the adjacent lane does not exist, exit the ADAS system and remind the driver to take over the vehicle control.
  • the position point corresponding to the preset distance on the lane line of the own lane, the position point corresponding to the CIPV, and the position point of the vehicle in the adjacent lane projected on the own lane it is then judged whether the first area is There is a hard boundary of the lane and whether there is an obstacle in the second area, wherein the steps of judging whether there is a hard boundary of the lane in the first area and judging whether there is an obstacle in the second area are performed in no particular order.
  • the corresponding position point offset processing is performed to obtain the offset position point.
  • the offset position point satisfies the preset conditions, then The offset position point is used as the control point at the current moment. If the offset position point does not meet the preset conditions, the ADAS system is exited and the driver is reminded to take over the vehicle control. When there is no hard boundary of the lane in the first area and there is no obstacle in the second area, it is judged whether the position point satisfies the preset conditions, and if the position point satisfies the preset conditions, the position point is regarded as the current moment If the position point does not meet the preset conditions, the ADAS system will be exited and the driver will be reminded to take over the control of the vehicle.
  • the control trajectory can be generated according to the determined control points, and the characteristics such as LCC and LKA can be realized.
  • the control trajectory can be generated according to the determined control points, which can realize the characteristics of TJA and OBF. , LKA and other characteristics; when any vehicle in the lane line, CIPV, and adjacent lane of the own lane exists, and there is a hard boundary of the lane in the first area, the control trajectory can be generated according to the determined control point, and the road edge can be realized. Away from such characteristics.
  • the control trajectory can be generated according to the determined control point, and characteristics such as CAA can be realized. In this way, different characteristics can be realized and multiple Features and smooth transitions between scenes.
  • FIG. 9 shows a schematic diagram of implementing an OBF characteristic by using a control point according to an embodiment of the present application.
  • the OBF characteristic requires that the self-vehicle can follow the vehicle ahead when there is no lane line and only the vehicle in front of the self-vehicle. track driving.
  • the lane line of the own lane does not exist in the environmental information at the current moment, and the CIPV 1101 exists.
  • the middle point 1102 of the CIPV rear bumper is selected as the position point.
  • the position point that is, the midpoint 1102 of the CIPV rear bumper, is used as the control point, and then the control trajectory at the current moment is generated to realize the OBF characteristic of the own vehicle 401 .
  • FIG. 10 shows a schematic diagram of implementing LCC characteristics using control points according to an embodiment of the present application.
  • the lane line of the own lane exists in the environmental information at the current moment, and the own lane is selected at this time.
  • the points 15-40m away from the vehicle in the longitudinal direction on the two lane lines of the own lane are obtained, so as to obtain the point 1202 that is 15-40m away from the vehicle in the longitudinal direction on the lane center line of the own lane, and this point is used as the position point.
  • the first area If there is no hard boundary of the lane, there is no obstacle in the second area, and the position point satisfies the preset conditions, the position point is used as the control point, and then the control trajectory at the current moment is generated to realize the LCC characteristics of the own vehicle 401 .
  • FIG. 11 shows a schematic diagram of implementing a TJA characteristic using control points according to an embodiment of the present application.
  • the TJA characteristic requires a lane line, but the sensor of the ego vehicle cannot accurately perceive the lane due to the occlusion of the preceding vehicle.
  • the position and shape information of the line that is, when the lane line data sensed by the sensor cannot meet the stability and accuracy requirements of the LCC characteristics, the vehicle in front is used for lateral control; at the same time, TJA cannot control the vehicle to completely follow the trajectory of the preceding vehicle.
  • the position point 1303 is used as the control point, and then the control trajectory at the current moment is generated, so as to ensure that when the CIPV deviates in the own lane or has a cut-in and cut-out action, the own vehicle can still be stable. Keep it centered and realize the TJA characteristics of the self-vehicle 401.
  • FIG. 12 shows a schematic diagram of implementing ALC/SLC characteristics by using control points according to an embodiment of the present application, and implementing ALC (as shown in FIGS. 12(a)-12(b) ) and SLC (as shown in FIG. 12(b) )
  • the essence of the characteristic process shown in 12(c) is the process of gradually shifting the control point in the own lane to the lane centerline of the adjacent lane.
  • FIG. 12(a) in response to the driver's lane change instruction, in the process of changing lanes, it is determined that the lane line of the own lane exists in the environmental information at the current moment, and at this time, the lane line of the own lane is selected in the longitudinal direction.
  • the position point is used as the control point, Then, the control trajectory at the current time is generated to realize the SLC characteristics of the own vehicle 401 .
  • a lane change is performed.
  • the point on the lane line of the lane is vertically distanced from the vehicle 401 by a preset distance, and the point is offset to the adjacent lane in the lateral direction by a certain distance to obtain the position point.
  • there are obstacles 1405 and 1404 in the second area then The position point offset processing is performed, and the offset position point 1406 that satisfies the preset condition is used as the control point, and then the control trajectory at the current moment is generated to realize the ALC characteristic of the own vehicle 401 .
  • Step 503 Generate a control trajectory at the current moment according to the control point at the current moment, the control trajectory at the historical moment, and the status information (such as motion status information) of the vehicle at the current moment, and the control trajectory represents the trajectory for guiding the vehicle to travel.
  • the control trajectory at the historical moment may include the control trajectory generated by the control point at the previous moment. It can be understood that, along with the real-time generation of the control trajectory, the control trajectory at the historical moment can be obtained; the control trajectory at the historical moment can represent the control trajectory in the time domain. The position of the control points at different times, in this way, the control trajectory is generated by using the continuity of the control points in space in the time domain, that is, only one point in space is needed to generate the control curve in the time dimension. The dimension of the object is reduced to one point. Compared with the method of generating a control trajectory by taking multiple points in the spatial domain, the embodiment of the present application does not need to generate multiple trajectory lines in advance, which greatly reduces the requirement for computing power and effectively improves ADAS. system performance.
  • FIG. 13 shows a comparison diagram of an intelligent driving according to an embodiment of the present application.
  • the real trajectory of the target vehicle 1505 is 1501;
  • 1503 is the control trajectory of the OBF characteristics generated at the time t1 through the control point at time t1, the control trajectory before the time t1 and the motion state information at the time t1, and 1104 is the control trajectory through the time t2
  • the trajectory 1501 is generated by fitting multiple points in the trajectory 1501 in the spatial dimension, while the trajectories 1503 and 1504 are the control points determined by the target vehicle 1505 at each time, and are generated by fitting in the time dimension based on the control points at different times.
  • control trajectory can be a third-order curve.
  • a third-order curve is generated by using the control point at the current moment, the control trajectory at the historical moment, and the motion state information at the current moment, and the third-order curve can be adapted to some ADAS system architectures. so as to improve the scope of application of the intelligent driving method in this application.
  • the control trajectory at the current moment may be generated by using a filter according to the control point at the current moment, the control trajectory at the historical moment, and the vehicle state information at the current moment.
  • the filter can be a Kalman filter (kalman filter), and the Kalman filter has time continuity, and the position information of the control point at the current moment, the control trajectory at the historical moment and the status information of the vehicle at the current moment are input.
  • the continuity of the control points in space in the time domain is used to generate the control trajectory at the current moment, so as to achieve different characteristics and improve the system performance. It is understandable that, with the passage of time, the control trajectories at different times are continuously generated, and the control trajectories at historical moments are continuously iterated.
  • the position information of the middle point 1102 of the CIPV rear bumper in the above-mentioned FIG. 10 , the control trajectory at the historical moment and the speed of the vehicle at the current moment, the yaw angular velocity, the steering radius and other information can be input into the Kalman filter to generate the current The control trajectory at the time, and use the control trajectory to track the vehicle, so as to realize the OBF characteristic.
  • the position information of the point 1202 on the lane center line of the own lane in the above-mentioned FIG. 11 which is 20 m away from the own vehicle in the longitudinal direction, the control trajectory at the historical moment, and the speed of the own vehicle at the current moment, the yaw rate, the turning radius, etc.
  • the information is input to the Kalman filter to generate the control trajectory at the current moment, and use the control trajectory to control the lane centering, so as to realize the LCC characteristic.
  • the midpoint 1302 of the CIPV rear bumper in the above-mentioned FIG. 12 can be projected onto the position information of the point 1303 on the center line of the lane of the own lane, the control trajectory at the historical moment, the speed of the vehicle at the current moment, and the yaw rate. , steering radius and other information are input into the Kalman filter to generate the control trajectory at the current moment, so as to realize the TJA characteristic.
  • the point on the lane centerline is used to generate the control trajectory, which ensures the smoothness of the control in the TJA characteristic.
  • the position information of the control point at the current moment in the lane change process in Figure 13 can be input into the Kalman filter, The control trajectory at the current moment is generated, and the control trajectory is used to perform lane change control, thereby realizing the ALC/SLC characteristics. Further, in the process of realizing the ALC/SLC characteristics, in order to ensure that the lane changing process conforms to the driver's habits, by analyzing a large number of driver's lane-changing data, the lateral offset speed curves corresponding to different lane-changing durations when changing lanes are obtained in advance.
  • the lateral deviation speed curve represents the lateral speed change at each moment when the own vehicle changes lanes.
  • FIG. 14 shows a schematic diagram of a lateral deviation speed curve according to an embodiment of the present application.
  • the driver changes The lateral speed change during lane conforms to a V-shaped curve, and for different lane-changing durations, it will concave or protrude on the basis of the V-shape.
  • the corresponding lateral offset velocity curve can be selected according to the preset or estimated lane change duration, and the corresponding lateral velocity at each moment can be determined according to the lateral offset velocity curve, thereby generating the control trajectory at each moment. , to achieve a comfortable lane change effect.
  • the environmental information around the vehicle at the current moment and the status information of the vehicle at the current moment are obtained; the control point at the current moment is determined according to the environmental information and the status information; the control point and the status information at the historical moment are determined according to the control point.
  • generate the control trajectory at the current moment and the control trajectory represents the trajectory used to guide the ego vehicle. In this way, based on the driver's driving habits and way of thinking, the driver usually focuses his attention on a point in the space at the current moment, which is the position the driver expects the vehicle to travel to.
  • the priority and corresponding weight information realize the switching between different features, which ensures a smooth transition between multiple features or scenarios, and does not need to use a state machine to switch between different features, reducing the logic complexity.
  • the embodiments of the present application further provide an intelligent driving device, and the intelligent driving device is used for implementing the technical solutions described in the above method embodiments.
  • FIG. 15 shows a schematic structural diagram of an intelligent driving device according to an embodiment of the present application; as shown in FIG. 15 , the intelligent driving device may include: an acquisition module 1701 for acquiring the environmental information around the vehicle at the current moment and the current The state information of the vehicle at the time; the determining module 1702 is used to determine the control point at the current time according to the environmental information and the state information; the generating module 1703 is used to determine the control point at the current time according to the control point, the control track at the historical time and the state information , to generate a control trajectory at the current moment, where the control trajectory represents a trajectory used to guide the ego vehicle to travel.
  • an acquisition module 1701 for acquiring the environmental information around the vehicle at the current moment and the current The state information of the vehicle at the time
  • the determining module 1702 is used to determine the control point at the current time according to the environmental information and the state information
  • the generating module 1703 is used to determine the control point at the current time according to the control point, the control track at the historical time and
  • the determining module 1702 is further configured to: sequentially determine the control point at the current moment according to the preset priority information and the status information of each first object in the environmental information .
  • the first object includes at least one of a lane line of the own lane of the own vehicle, a target vehicle, and a vehicle in an adjacent lane of the own vehicle; wherein the lane of the own vehicle The priority of the line, the target vehicle, and the vehicles in the adjacent lanes decreases in sequence;
  • the determining module 1702 is further configured to: in the case that the lane line of the own lane exists, use the position point corresponding to the preset distance on the lane line of the own lane and meeting the preset condition as the current moment. control point; the preset distance is determined based on the state information; if the lane line of the own lane does not exist, in the case that the target vehicle exists, the target vehicle corresponds to and meets the preset condition The position point is used as the control point at the current moment; if the lane line of the own lane does not exist, and the target vehicle does not exist, then if the vehicle in the adjacent lane exists, the vehicle in the adjacent lane will be placed in the adjacent lane. The position point projected from the lane and meeting the preset conditions is used as the control point at the current moment.
  • the determining module 1702 is further configured to: obtain the position point corresponding to the first object with the highest priority; determine whether there is a second object in the preset area, and the second object is the environment The avoidance object of the own vehicle in the information; when the second object exists, offset processing is performed on the position point corresponding to the first object, and the offset position point is used as the control point at the current moment.
  • the second object includes a hard boundary of the own lane;
  • the preset area is an area with a preset size centered on a position point corresponding to the first object;
  • the determining module 1702 is further configured to: when the hard boundary of the self-lane exists, perform offset processing on the position point corresponding to the first object, so that the hard boundary can be changed with the offset after the hard boundary.
  • the position point is outside the area of the center preset size, and the offset position point is within the own lane; the offset position point is used as the control point at the current moment.
  • the second object includes an obstacle;
  • the preset area is an area of a preset size in front of the vehicle in the vehicle lane;
  • the determining module 1702 is further configured to: when the obstacle exists, perform offset processing on the position point corresponding to the first object according to the weight information corresponding to the obstacle, so that the obstacle is The position point after the offset is outside the area of the preset size, and the position point after the offset is in the own lane; wherein, the weight information represents the degree of the vehicle avoiding the obstacle; the The offset position point is used as the control point at the current moment.
  • the value of the weight information is determined by at least one of the headway THW, the time to collision TTC, and the intrusion width IVW, wherein the value of the weight information is negatively correlated with the THW, It is negatively correlated with the TTC and positively correlated with the IVW.
  • the generating module 1703 is further configured to: generate the control trajectory at the current moment by using a Kalman filter according to the control point, the control trajectory at the historical moment, and the state information.
  • the environmental information around the vehicle at the current moment and the status information of the vehicle at the current moment are obtained; the control point at the current moment is determined according to the environmental information and the status information; according to the control point, the control trajectory and the status information at the historical moment , generate the control trajectory at the current moment, and the control trajectory represents the trajectory used to guide the ego vehicle.
  • the driver usually focuses his attention on a point in the space at the current moment, which is the position the driver expects the vehicle to travel to.
  • the way of prioritization change the position of the control point, determine the control point at the current moment, different first objects correspond to different characteristics, so as to realize multiple characteristics (such as: OBF, LKA, TJA, LCC, etc.),
  • different first objects correspond to different characteristics, so as to realize multiple characteristics (such as: OBF, LKA, TJA, LCC, etc.)
  • switching between different features is realized, which ensures a smooth transition between multiple features or scenarios, and does not need to use a state machine to switch between different features, reducing the need for logic complexity.
  • FIG. 16 shows an ADAS system architecture diagram according to an embodiment of the present application.
  • the ADAS system architecture includes: a preprocessing layer, a planning layer, a decision-making layer, a control layer, etc., and may also include an execution layer and the like.
  • the preprocessing layer is used to generate the environmental information around the self-vehicle at the current moment; exemplarily, the data detected by the sensors installed on the self-vehicle can be processed to obtain the environmental information around the self-vehicle.
  • map information can also be obtained based on GPS data, high-precision map data, etc., for example, the distance between the vehicle and the intersection ahead, the lane where the vehicle is located, and so on.
  • Planning layer used to determine the control point at the current moment according to the environmental information and the status information of the vehicle at the current moment; generate the control trajectory at the current moment according to the control point, the control trajectory at the historical moment and the status information;
  • the control trajectory represents the trajectory used to guide the driving of the self-vehicle;
  • this layer takes the environmental information obtained by the preprocessing layer as input, combined with the state information of the self-vehicle (which may include: driving and riding state information and self-vehicle state information). information), through the intelligent driving device, the intelligent driving method of the present application is executed, so as to determine the characteristics (or functions) applicable to the current scene and the control trajectory corresponding to the characteristics, and input the control trajectory to the decision-making layer.
  • the planning layer may also sequentially determine the control point at the current moment according to the preset priority information of each first object in the environmental information and the state information. It is also possible to obtain the position point corresponding to the first object with the highest priority; determine whether there is a second object in the preset area, and the second object is the avoidance object of the own vehicle in the environmental information; when the second object exists, Offset processing is performed on the position point corresponding to the first object, and the offset position point is used as the control point at the current moment.
  • Decision layer used to judge whether the control trajectory is suitable for the working state of the vehicle at the current moment; exemplarily, this layer takes the control trajectory generated by the planning layer as input, , engine, electric power steering and other systems), judge whether the control trajectory can be applied to the current working state of the vehicle, when it is judged that the control trajectory can be applied to the current working state of the vehicle, enter the control trajectory to the control layer.
  • Control layer used to generate a control signal at the current moment based on the control trajectory when the control trajectory is applicable to the working state of the vehicle at the current moment, and the control signal is used for assisted driving of the vehicle; exemplarily , this layer takes the control trajectory of the decision-making layer as the input, and combines the dynamic information of the vehicle to generate control signals such as steering wheel angle and acceleration.
  • system may also include an execution layer: taking the control signal generated by the control layer as an input, executing the control signal, for example, executing the control signal through an electric power steering system (Electric Power Steering, EPS) and an engine, thereby realizing the vehicle assisted driving.
  • EPS Electric Power Steering
  • the planning layer through the planning layer, the influence of each object in the environmental information on the characteristics is concentrated on the point in space, the point is used as the control point, and the control trajectory is generated through the continuity of the control point in the time dimension. , so as to achieve different characteristics. In this way, only the trajectory generator used to generate the control trajectory through the control point is needed, and multiple trajectory generators are not needed, which greatly reduces the computing power requirement and improves the system performance.
  • the planning layer is determined by the preset priorities of different objects in the environmental information and the corresponding weight information. Control points at each moment to achieve smooth transition of different characteristics and scenes.
  • ADAS system if there is a new feature, only the processing of the object corresponding to the feature needs to be added in the preprocessing layer, and the priority and weight information of the object corresponding to the new feature can also be set.
  • the newly generated trajectory line for this characteristic does not need to adjust the state machine, and the system has high scalability.
  • An embodiment of the present application provides an intelligent driving device, comprising: a processor and a memory for storing instructions executable by the processor; wherein the processor is configured to implement the above intelligent driving method when executing the instructions.
  • FIG. 17 shows a schematic structural diagram of another intelligent driving device according to an embodiment of the present application.
  • the intelligent driving device may include: at least one processor 1801 , a communication line 1802 , a memory 1803 and at least one communication Interface 1804.
  • the processor 1801 may be a general-purpose central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more processors used to control the execution of the programs of the present application. integrated circuit.
  • CPU central processing unit
  • ASIC application-specific integrated circuit
  • Communication line 1802 may include a path to communicate information between the components described above.
  • the communication interface 1804 using any transceiver-like device, is used to communicate with other devices or communication networks, such as Ethernet, RAN, wireless local area networks (WLAN), etc.
  • devices or communication networks such as Ethernet, RAN, wireless local area networks (WLAN), etc.
  • Memory 1803 may be read-only memory (ROM) or other types of static storage devices that can store static information and instructions, random access memory (RAM) or other types of information and instructions It can also be an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disk storage, CD-ROM storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or capable of carrying or storing desired program code in the form of instructions or data structures and capable of being executed by a computer Access any other medium without limitation.
  • the memory may exist independently and be connected to the processor through communication line 1802. The memory can also be integrated with the processor.
  • the memory provided by the embodiments of the present application may generally be non-volatile.
  • the memory 1803 is used for storing computer-executed instructions for executing the solution of the present application, and the execution is controlled by the processor 1801 .
  • the processor 1801 is configured to execute the computer-executed instructions stored in the memory 1803, thereby implementing the methods provided in the foregoing embodiments of the present application.
  • the computer-executed instructions in this embodiment of the present application may also be referred to as application code.
  • the processor 1801 may include one or more CPUs, such as CPU0 and CPU1 in FIG. 17 .
  • the intelligent driving device may include multiple processors, for example, the processor 1801 and the processor 1807 in FIG. 17 .
  • processors can be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor.
  • a processor herein may refer to one or more devices, circuits, and/or processing cores for processing data (eg, computer program instructions).
  • the intelligent driving apparatus may further include an output device 1805 and an input device 1806 .
  • the output device 1805 is in communication with the processor 1801 and can display information in a variety of ways.
  • the output device 1805 may be a liquid crystal display (LCD), a light emitting diode (LED) display device, a cathode ray tube (CRT) display device, or a projector (projector) Wait.
  • Input device 1806 is in communication with processor 1801 and can receive user input in a variety of ways.
  • the input device 1806 may be a mouse, a keyboard, a touch screen device or a sensing device, or the like.
  • the embodiments of the present application further provide an advanced driving assistance system, which is applied to unmanned driving or intelligent driving.
  • the sensor is used for sensing environmental information, and at least one device in the system can be integrated into a complete machine or equipment, or at least one device in the system can also be independently set as a component or device.
  • An embodiment of the present application further provides a vehicle, where the vehicle includes at least one intelligent driving device or any of the above-mentioned systems mentioned in the above-mentioned embodiments of the present application.
  • Embodiments of the present application provide a non-volatile computer-readable storage medium on which computer program instructions are stored, and when the computer program instructions are executed by a processor, implement the above method.
  • Embodiments of the present application provide a computer program product, including computer-readable codes, or a non-volatile computer-readable storage medium carrying computer-readable codes, when the computer-readable codes are stored in a processor of an electronic device When running in the electronic device, the processor in the electronic device executes the above method.
  • a computer-readable storage medium may be a tangible device that can hold and store instructions for use by the instruction execution device.
  • the computer-readable storage medium may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing.
  • Computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read only memory (ROM), erasable programmable read-only memory (Electrically Programmable Read-Only-Memory, EPROM or flash memory), static random access memory (Static Random-Access Memory, SRAM), portable compact disk read-only memory (Compact Disc Read-Only Memory, CD - ROM), Digital Video Disc (DVD), memory sticks, floppy disks, mechanically encoded devices, such as punch cards or raised structures in grooves on which instructions are stored, and any suitable combination of the foregoing .
  • RAM random access memory
  • ROM read only memory
  • EPROM erasable programmable read-only memory
  • EPROM Errically Programmable Read-Only-Memory
  • SRAM static random access memory
  • portable compact disk read-only memory Compact Disc Read-Only Memory
  • CD - ROM Compact Disc Read-Only Memory
  • DVD Digital Video Disc
  • memory sticks floppy disks
  • Computer readable program instructions or code described herein may be downloaded to various computing/processing devices from a computer readable storage medium, or to an external computer or external storage device over a network such as the Internet, a local area network, a wide area network and/or a wireless network.
  • the network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and/or edge servers.
  • a network adapter card or network interface in each computing/processing device receives computer-readable program instructions from a network and forwards the computer-readable program instructions for storage in a computer-readable storage medium in each computing/processing device .
  • the computer program instructions used to perform the operations of the present application may be assembly instructions, Instruction Set Architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, or in one or more source or object code written in any combination of programming languages, including object-oriented programming languages such as Smalltalk, C++, etc., and conventional procedural programming languages such as the "C" language or similar programming languages.
  • the computer readable program instructions may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer, or entirely on the remote computer or server implement.
  • the remote computer may be connected to the user's computer through any kind of network—including a Local Area Network (LAN) or a Wide Area Network (WAN)—or, may be connected to an external computer (eg, use an internet service provider to connect via the internet).
  • electronic circuits such as programmable logic circuits, Field-Programmable Gate Arrays (FPGA), or Programmable Logic Arrays (Programmable Logic Arrays), are personalized by utilizing state information of computer-readable program instructions.
  • Logic Array, PLA the electronic circuit can execute computer readable program instructions to implement various aspects of the present application.
  • These computer readable program instructions may be provided to a processor of a general purpose computer, special purpose computer or other programmable data processing apparatus to produce a machine that causes the instructions when executed by the processor of the computer or other programmable data processing apparatus , resulting in means for implementing the functions/acts specified in one or more blocks of the flowchart and/or block diagrams.
  • These computer readable program instructions can also be stored in a computer readable storage medium, these instructions cause a computer, programmable data processing apparatus and/or other equipment to operate in a specific manner, so that the computer readable medium on which the instructions are stored includes An article of manufacture comprising instructions for implementing various aspects of the functions/acts specified in one or more blocks of the flowchart and/or block diagrams.
  • Computer readable program instructions can also be loaded onto a computer, other programmable data processing apparatus, or other equipment to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other equipment to produce a computer-implemented process , thereby causing instructions executing on a computer, other programmable data processing apparatus, or other device to implement the functions/acts specified in one or more blocks of the flowcharts and/or block diagrams.
  • each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more functions for implementing the specified logical function(s) executable instructions.
  • the functions noted in the blocks may occur out of the order noted in the figures. For example, two blocks in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved.
  • each block of the block diagrams and/or flowchart illustrations, and combinations of blocks in the block diagrams and/or flowchart illustrations can be implemented in hardware (eg, circuits or ASICs (Application) that perform the corresponding functions or actions. Specific Integrated Circuit, application-specific integrated circuit)), or can be implemented by a combination of hardware and software, such as firmware.

Landscapes

  • Engineering & Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Transportation (AREA)
  • Mechanical Engineering (AREA)
  • Human Computer Interaction (AREA)
  • Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Traffic Control Systems (AREA)
  • Control Of Driving Devices And Active Controlling Of Vehicle (AREA)

Abstract

一种智能驾驶方法、装置、存储介质及计算机程序，可用于高级辅助驾驶和自动驾驶，该方法包括：获取当前时刻自车周围的环境信息及当前时刻自车的状态信息；根据环境信息及状态信息，确定当前时刻的控制点；根据控制点、历史时刻的控制轨迹及状态信息，生成当前时刻的控制轨迹，控制轨迹表示用于导引自车行驶的轨迹。这样，将环境信息对特性的影响集中到控制点，通过控制点生成当前时刻的控制轨迹，满足不同场景下的控制要求，且无需预先对环境信息进行处理生成多个条轨迹线，大大降低了对算力的要求，根据多个信息源的优先级及对应的权重信息，实现不同特性之间的切换，保证了多个特性或场景间的平滑过渡，提升了高级驾驶辅助系统的性能。

Description

一种智能驾驶方法、装置、存储介质及计算机程序 技术领域
本申请涉及自动驾驶技术领域，特别涉及一种智能驾驶方法、装置、存储介质及计算机程序。
背景技术
在高级驾驶辅助系统(Advanced Driver Assistant，ADAS)中，通过车辆的传感器感知、融合得到车辆周边的环境信息，进而根据环境信息进行路径规划、车辆控制等，实现辅助驾驶员驾驶。
目前，ADAS中所具有的特性(又称功能)可包括以下中的一种或多种：自适应巡航(Adaptive Cruise Control，ACC)、车道保持辅助(Lane Keep Assistance，LKA)、交通拥堵辅助(Traffic Jam Assistance，TJA)、自动变道(Auto Lane Change，ALC)、半自动变道(Semi-Auto Lane Change，SLC)、智慧闪躲(Collision Avoid Assistance，CAA)、车道居中控制(Lane Center Control，LCC)、目标跟随(Object Follow，OBF)等等。在一些ADAS系统的架构中，每个特性对应的控制轨迹基于多条轨迹线生成，为了支持不同特性，需要对环境信息进行处理，生成多条轨迹线，对ADAS系统的算力要求较高；同时，利用状态机实现上述不同特性之间的切换，切换条件与过渡逻辑较复杂；另外，系统的可扩展性差，若新增特性，需要新生成针对该特性的轨迹线，并需要对状态机进行调整。
发明内容
本申请提出了一种智能驾驶方法、装置、存储介质及计算机程序。
第一方面，本申请的实施例提供了一种智能驾驶方法，所述方法包括：获取当前时刻自车周围的环境信息及当前时刻自车的状态信息；根据所述环境信息及所述状态信息，确定当前时刻的控制点；根据所述控制点、历史时刻的控制轨迹及所述状态信息，生成当前时刻的控制轨迹，所述控制轨迹表示用于导引自车行驶的轨迹。
基于上述技术方案，可以根据当前时刻自车周围的环境信息及当前时刻自车的状态信息，确定当前时刻的控制点；并根据该控制点、历史时刻的控制轨迹及当前时刻自车的状态信息，生成当前时刻的控制轨迹；这样，将环境信息对特性的影响集中到空间中的点(即控制点)，通过该控制点即可生成当前时刻的控制轨迹，满足不同场景下的控制要求，且无需预先对环境信息进行处理生成多个条轨迹线，大大降低了对算力的要求，从而提升了ADAS系统的性能。
根据第一方面，在所述第一方面的第一种可能的实现方式中，所述根据所述环境信息及所述状态信息，确定当前时刻的控制点，包括：根据所述环境信息中各第一对象的预设优先级信息及所述状态信息，依次确定所述当前时刻的控制点。
基于上述技术方案，根据环境信息中各第一对象的预设优先级信息及状态信息，依次 确定当前时刻的控制点，这样，通过多个信息源(即环境信息中各第一对象)，按照各信息源的优先级排序的方式，改变不同场景的控制点的位置，确定当前时刻的控制点，从而实现多个特性(如：OBF、LKA、TJA、LCC等特性)，同时，根据多个信息源的优先级，实现不同特性之间的切换，保证了多个特性或场景间的平滑过渡，且无需利用状态机实现不同特性之间的切换，降低了逻辑复杂度。
根据第一方面的第一种可能的实现方式，在所述第一方面的第二种可能的实现方式中，所述第一对象至少包括：自车道的车道线、目标车辆、邻车道中车辆的一个；其中，所述自车道的车道线、所述目标车辆、所述邻车道中车辆的优先级依次降低；所述根据所述环境信息中各第一对象的预设优先级信息及所述状态信息，依次确定所述当前时刻的控制点，包括：在所述自车道的车道线存在的情况下，将所述自车道的车道线上预设距离对应的、且符合预设条件的位置点作为当前时刻的控制点，所述预设距离基于所述状态信息确定；如果所述自车道的车道线不存在，则在所述目标车辆存在的情况下，将所述目标车辆对应的、且符合预设条件的位置点作为当前时刻的控制点；如果所述自车道的车道线不存在，且所述目标车辆不存在，则在所述邻车道中车辆存在的情况下，将所述邻车道中车辆在自车道投影的、且符合预设条件的位置点作为当前时刻的控制点。
基于上述技术方案，可以根据当前时刻自车周围的环境信息中自车的自车道的车道线确定当前时刻的控制点，进而实现LCC、LKA等特性；可以根据当前时刻自车周围的环境信息中CIPV，确定当前时刻的控制点，进而实现TJA、OBF等特性；可以根据当前时刻自车周围的环境信息中邻车道中车辆，确定当前时刻的控制点，进而实现TJA、LKA等特性；这样，利用自车道的车道线、目标车辆、邻车道中车辆，按照优先级排序的方式，改变不同场景的控制点的位置，根据环境信息中存在的、优先级最高的第一对象，确定当前时刻的控制点，从而实现多个特性(如：LCC、LKA、OBF、TJA等特性)，同时，根据自车道的车道线、目标车辆、邻车道中车辆的优先级，实现不同特性之间的切换，保证了多个特性或场景间的平滑过渡。
根据第一方面的第一种或第二种可能的实现方式，在所述第一方面的第三种可能的实现方式中，所述根据所述环境信息中各第一对象的预设优先级信息及所述状态信息，依次确定所述当前时刻的控制点，包括：获取优先级最高的第一对象对应的位置点；判断预设区域内是否存在第二对象，所述第二对象为环境信息中自车的避让对象；在所述第二对象存在时，对所述第一对象对应的位置点进行偏移处理，并将偏移后的位置点作为当前时刻的控制点。
基于上述技术方案，在预设区域中存在第二对象时，说明若以上述获取的第一对象对应的位置点生成控制轨迹，并按照该控制轨迹控制自车行驶，存在碰撞第二对象的风险，此时，对第一对象对应的位置点进行偏移处理，并将偏移后的位置点作为当前时刻的控制点，从而避免碰撞到第二对象，这样，基于第一对象，按照优先级排序的方式，并结合存在的第二对象，共同改变不同场景的控制点的位置，确定当前时刻的控制点，可以实现多个特性(如CAA、路沿远离等特性)，并保证了多个特性或场景间的平滑过渡。
根据第一方面的第三种可能的实现方式，在所述第一方面的第四种可能的实现方式中，所述第二对象包括自车道的硬边界；所述预设区域为以第一对象对应的位置点为中心预设大小的区域；所述在所述第二对象存在时，对所述第一对象对应的位置点进行偏移处理， 并将偏移后的位置点作为当前时刻的控制点，包括：在所述自车道的硬边界存在时，对所述第一对象对应的位置点进行偏移处理，以使所述硬边界在以所述偏移后的位置点为中心预设大小的区域之外，且所述偏移后的位置点在自车道内；将偏移后的位置点作为当前时刻的控制点。
基于上述技术方案，在自车道的硬边界存在时，对第一对象对应的位置点进行偏移处理，以使该硬边界在以偏移后的位置点为中心预设大小的区域之外，且偏移后的位置点在自车道内，并将偏移后的位置点作为当前时刻的控制点，该控制点远离车道的硬边界，从而使得自车根据该控制点生成的控制轨迹行驶时，避免碰撞到硬边界，这样，基于第一对象，按照优先级排序的方式，并结合硬边界，共同改变不同场景的控制点的位置，确定当前时刻的控制点，可以实现多个特性(如路沿远离等特性)，并保证了多个特性或场景间的平滑过渡。
根据第一方面的第三种可能的实现方式，在所述第一方面的第五种可能的实现方式中，所述第二对象包括障碍物；所述预设区域为所述自车道中自车前方预设大小的区域；所述在所述第二对象存在时，对所述第一对象对应的位置点进行偏移处理，并将偏移后的位置点作为当前时刻的控制点，包括：在所述障碍物存在时，根据所述障碍物对应的权重信息，对所述第一对象对应的位置点进行偏移处理，以使所述障碍物在以所述偏移后的位置点为中心预设大小的区域之外，且所述偏移后的位置点在自车道内；其中，所述权重信息表示自车避让该障碍物的程度；将偏移后的位置点作为当前时刻的控制点。
基于上述技术方案，考虑到不同的场景中，自车避让障碍物的动作不同，利用障碍物对应的权重信息表示不同场景下自车避让该障碍物的程度，并根据障碍物对应的权重信息，对第一对象对应的位置点进行偏移处理，从而实现不同场景下的位置点偏移，将偏移后的位置点作为当前时刻的控制点，该控制点远离障碍物，从而使得自车案根据该控制点生成的控制轨迹行驶时，避免碰撞到该障碍物，这样，基于第一对象按照优先级排序的方式，并结合障碍物及障碍物对应的权重信息，共同改变不同场景的控制点的位置，确定当前时刻的控制点，从而实现多个特征(如CAA等特性)，并保证了多个特性或场景间的平滑过渡。
根据第一方面的第五种可能的实现方式，在所述第一方面的第六种可能的实现方式中，所述权重信息的数值至少由车头时距THW、碰撞时间TTC、侵入宽度IVW中的一个确定，其中，所述权重信息的数值与所述THW成负相关，与所述TTC成负相关，与所述IVW成正相关。
基于上述技术方案，权重信息的数值与THW成负相关，即THW越小，自车避让障碍物的动作越明显，权重信息的数值与TTC成负相关，即TTC越小，自车避让障碍物的动作越明显，权重信息的数值与IVW成正相关，即IVW越大，自车避让障碍物的动作越明显；这样，根据不同场景下对应的权重信息的数值，改变不同场景下控制点的位置，确定当前时刻的控制点，实现了不同场景的平滑过渡。
根据第一方面，在所述第一方面的第七种可能的实现方式中，所述根据所述控制点、历史时刻的控制轨迹及所述状态信息，生成当前时刻的控制轨迹，包括：根据所述控制点、历史时刻的控制轨迹及所述状态信息，利用卡尔曼滤波器，生成所述当前时刻的控制轨迹。
基于上述技术方案，卡尔曼滤波器具有时间连续性，将当前时刻控制点的位置信息、 历史时刻的控制轨迹及当前时刻自车状态信息，输入到卡尔曼滤波器，利用空间上的控制点在时间域上的连续性生成当前时刻的控制轨迹，满足不同场景下的控制要求，且大大降低了对算力的要求，提高了系统性能。
第二方面，本申请的实施例提供了一种智能驾驶装置，所述装置包括：获取模块，用于获取当前时刻自车周围的环境信息及当前时刻自车的状态信息；确定模块，用于根据所述环境信息及所述状态信息，确定当前时刻的控制点；生成模块，用于根据所述控制点、历史时刻的控制轨迹及所述状态信息，生成当前时刻的控制轨迹，所述控制轨迹表示用于导引自车行驶的轨迹。
根据第二方面，在所述第二方面的第一种可能的实现方式中，所述确定模块，还用于：根据所述环境信息中各第一对象的预设优先级信息及所述状态信息，依次确定所述当前时刻的控制点。
根据第二方面的第一种可能的实现方式，在所述第二方面的第二种可能的实现方式中，所述第一对象至少包括所述自车道的车道线、目标车辆、所述邻车道中车辆的一个；其中，所述自车道的车道线、所述目标车辆、所述邻车道中车辆的优先级依次降低；所述确定模块，还用于：在所述自车道的车道线存在的情况下，将所述自车道的车道线上预设距离对应的、且符合预设条件的位置点作为当前时刻的控制点；所述预设距离基于所述状态信息确定；如果所述自车道的车道线不存在，则在所述目标车辆存在的情况下，将所述目标车辆对应的、且符合预设条件的位置点作为当前时刻的控制点；如果所述自车道的车道线不存在，且所述目标车辆不存在，则在所述邻车道中车辆存在的情况下，将所述邻车道中车辆在自车道投影的、且符合预设条件的位置点作为当前时刻的控制点。
根据第二方面的第一种或第二种可能的实现方式，在所述第二方面的第三种可能的实现方式中，所述确定模块，还用于：获取优先级最高的第一对象对应的位置点；判断预设区域内是否存在第二对象，所述第二对象为环境信息中自车的避让对象；在所述第二对象存在时，对所述第一对象对应的位置点进行偏移处理，并将偏移后的位置点作为当前时刻的控制点。
根据第二方面的第三种可能的实现方式，在所述第二方面的第四种可能的实现方式中，所述第二对象包括自车道的硬边界；所述预设区域为以第一对象对应的位置点为中心预设大小的区域；所述确定模块，还用于：在所述自车道的硬边界存在时，对所述第一对象对应的位置点进行偏移处理，以使所述硬边界在以所述偏移后的位置点为中心预设大小的区域之外，且所述偏移后的位置点在自车道内；将偏移后的位置点作为当前时刻的控制点。
根据第二方面的第三种可能的实现方式，在所述第二方面的第五种可能的实现方式中，所述第二对象包括障碍物；所述预设区域为所述自车道中自车前方预设大小的区域；所述确定模块，还用于：在所述障碍物存在时，根据所述障碍物对应的权重信息，对所述第一对象对应的位置点进行偏移处理，以使所述障碍物在以所述偏移后的位置点为中心预设大小的区域之外，且所述偏移后的位置点在自车道内；其中，所述权重信息表示自车避让该障碍物的程度；将偏移后的位置点作为当前时刻的控制点。
根据第二方面的第五种可能的实现方式，在所述第二方面的第六种可能的实现方式中，所述权重信息的数值至少由车头时距THW、碰撞时间TTC、侵入宽度IVW中的一个确定，其中，所述权重信息的数值与所述THW成负相关，与所述TTC成负相关，与所述IVW成 正相关。
根据第二方面，在所述第二方面的第七种可能的实现方式中，所述生成模块，还用于：根据所述当前时刻的控制点、历史时刻的控制轨迹及所述状态信息，利用卡尔曼滤波器，生成所述当前时刻的控制轨迹。
上述第二方面及第二方面的各种可能的实现方式的技术效果，参见上述第一方面。
第三方面，本申请的实施例提供了一种智能驾驶装置，包括：处理器；用于存储处理器可执行指令的存储器；其中，所述处理器被配置为执行所述指令时实现上述第一方面或者第一方面的多种可能的实现方式中的一种或几种的智能驾驶方法。
上述第三方面的技术效果，参见上述第一方面。
第四方面，本申请的实施例提供了一种非易失性计算机可读存储介质，其上存储有计算机程序指令，所述计算机程序指令被处理器执行时实现上述第一方面或者第一方面的多种可能的实现方式中的一种或几种的智能驾驶方法。
上述第四方面的技术效果，参见上述第一方面。
第五方面，本申请的实施例提供了一种包含指令的计算机程序产品，当其在计算机上运行时，使得计算机执行上述第一方面或者第一方面的多种可能的实现方式中的一种或几种的智能驾驶方法。
上述第五方面的技术效果，参见上述第一方面。
第六方面，本申请实施例还提供一种车辆，所述车辆包括上述第二方面或者第二方面的多种可能的实现方式中的一种或几种的智能驾驶装置。
上述第六方面的技术效果，参见上述第一方面。
第七方面，本申请实施例还提供一种自动驾驶辅助系统，所述系统包括：预处理层、规划层、决策层、控制层；其中，所述预处理层，用于生成当前时刻自车周围的环境信息；所述规划层，用于根据所述环境信息及当前时刻自车的状态信息，确定当前时刻的控制点；根据所述控制点、历史时刻的控制轨迹及所述状态信息，生成当前时刻的控制轨迹；所述控制轨迹表示用于导引自车行驶的轨迹；所述决策层，用于对所述控制轨迹是否适用于当前时刻自车的工作状态进行判断；所述控制层，用于在所述控制轨迹适用于当前时刻自车的工作状态的情况下，基于所述控制轨迹，生成当前时刻的控制信号，所述控制信号用于车辆的辅助驾驶。
根据第七方面，在所述第七方面的第一种可能的实现方式中，所述规划层还用于：根据所述环境信息中各第一对象的预设优先级信息及所述状态信息，依次确定所述当前时刻的控制点。
根据第七方面的第一种可能的实现方式，在所述第七方面的第二种可能的实现方式中，所述规划层还用于：获取优先级最高的第一对象对应的位置点；判断预设区域内是否存在第二对象，所述第二对象为环境信息中自车的避让对象；在所述第二对象存在时，对所述第一对象对应的位置点进行偏移处理，并将偏移后的位置点作为当前时刻的控制点。
上述第七方面及第七方面的各种可能的实现方式的技术效果，参见上述第一方面，此外，基于第七方面，若有新增特性，仅需在系统的预处理层增加针对该特性对应的对象的处理，还可以设定新增特定对应对象的优先级等信息，不需要新生成针对该特性的轨迹线，也无需对状态机进行调整，系统可拓展性高。
本申请的这些和其他方面在以下(多个)实施例的描述中会更加简明易懂。
附图说明
图1示出根据本申请一实施例的智能驾驶方法适用的应用场景示意图；
图2示出根据本申请一实施例的一种智能驾驶方法的流程图；
图3示出根据本申请一实施例的一种确定当前时刻控制点的流程图；
图4示出根据本申请一实施例的多种车辆入侵自车道的示意图；
图5示出根据本申请一实施例的针对图4中多种车辆入侵自车道场景中位置点偏移的示意图；
图6示出根据本申请一实施例的一种权重体函数的示意图；
图7示出根据本申请一实施例的另一种权重体函数的示意图；
图8示出根据本申请一实施例的一种确定当前时刻控制点的流程图；
图9示出根据本申请一实施例的一种利用控制点实现OBF特性的示意图；
图10示出根据本申请一实施例的一种利用控制点实现LCC特性的示意图；
图11示出根据本申请一实施例的一种利用控制点实现TJA特性的示意图；
图12示出根据本申请一实施例的一种利用控制点实现ALC/SLC特性的示意图；
图13示出根据本申请一实施例的一种智能驾驶的对比图；
图14示出根据本申请一实施例的一种横向偏移速度曲线的示意图；
图15示出根据本申请一实施例的一种智能驾驶装置的结构示意图；
图16示出根据本申请一实施例的ADAS系统架构图；
图17示出根据本申请一实施例的另一种智能驾驶装置的结构示意图。
具体实施方式
以下将参考附图详细说明本申请的各种示例性实施例、特征和方面。附图中相同的附图标记表示功能相同或相似的元件。尽管在附图中示出了实施例的各种方面，但是除非特别指出，不必按比例绘制附图。
在这里专用的词“示例性”意为“用作例子、实施例或说明性”。这里作为“示例性”所说明的任何实施例不必解释为优于或好于其它实施例。
本申请提供了一种智能驾驶方法，该方法可以通过智能驾驶装置执行，其中，该智能驾驶装置可以是独立设置，也可以集成在其他装置中，还可以是通过软件或者软件与硬件结合实现。
示例性地，该智能驾驶装置可以应用于ADAS或自动驾驶系统(Automated Driving System，ADS)中，还可以应用于物物通信(Device to Device Communication，D2D)、车与任何事物相通信(vehicle to everything，V2X)、车与车通信(Vehicle to Vehicle，V2V)、长期演进与车通信(long term evolution-vehicle，LTE-V)、长期演进与机器通信(long term evolution-machine，LTE-M)等场景。
示例性地，该智能驾驶装置可为具有控制轨迹生成功能的车辆，或者为具有控制轨迹生成功能的其他部件。该智能驾驶装置包括但不限于：车载终端、车载控制器、车载模块、车载模组、车载部件、车载芯片、车载单元、车载雷达或车载摄像头等传感器，车辆可通 过该车载终端、车载控制器、车载模块、车载模组、车载部件、车载芯片、车载单元、车载雷达或摄像头，实施本申请提供的智能驾驶方法。
示例性地，该智能驾驶装置还可以为除了车辆之外的其他具有控制轨迹生成功能的智能终端，或设置在除了车辆之外的其他具有控制轨迹生成功能的智能终端中，或设置于该智能终端的部件中。该智能终端可以为智能运输设备、智能家居设备、机器人、无人机等其他终端设备。该智能驾驶装置包括但不限于智能终端或智能终端内的控制器、芯片、雷达或摄像头等传感器、以及其他部件等。
示例性地，该智能驾驶装置可以是一个通用设备或者是一个专用设备。在具体实现中，该智能驾驶装置还可以为台式机、便携式电脑、网络服务器、云服务器、掌上电脑(personal digital assistant，PDA)、移动手机、平板电脑、无线终端设备、嵌入式设备或其他具有处理功能的设备。
示例性地，该智能驾驶装置还可以是具有处理功能的芯片或处理器，该智能驾驶装置可以包括多个处理器。处理器可以是一个单核(single-CPU)处理器，也可以是一个多核(multi-CPU)处理器。该具有处理功能的芯片或处理器可以设置在传感器中，也可以不设置在传感器中，而设置在传感器输出信号的接收端。
为了便于描述，本申请实施例中，以智能驾驶装置应用于ADAS系统中为例，对提供的智能驾驶方法进行示例性地说明。
下面对本申请实施例提供的智能驾驶方法进行详细说明。
图1示出根据本申请一实施例的智能驾驶方法适用的应用场景示意图。如图1所示，该应用场景可以为高级辅助驾驶场景，该场景中包括：自车401，自车401中安装有至少一个传感器(图中未示出)，自车401还包括智能驾驶装置(图中未示出)，该传感器用于感知自车401周围的环境信息，例如，自车道(即自车所在的车道)的车道线403、邻车道(即与自车道相邻的车道)的车道线404、道路的路沿边界405、行人406、施工区域407、他车(如，前方车辆402、侧方车辆(图中未示出)、后方车辆(图中未示出))等等，该传感器将感知到的环境信息传输到智能驾驶装置，该智能驾驶装置用于执行本申请实施例中的智能驾驶方法。
需要说明的是，图1中仅以一辆自车、一辆他车、三个车道、一个行人及一个施工区域示出，应理解，应用场景中可以包括更多或更少的车辆、车道及障碍物等，此处不再示出。
此外，本申请实施例描述的应用场景是为了更加清楚的说明本申请实施例的技术方案，本领域技术人员可知，针对其他相似的或新的应用场景的出现，本申请实施例提供的技术方案对于类似的技术问题，同样适用。
图2示出根据本申请一实施例的一种智能驾驶方法的流程图。如图2所示，该方法可以应用于上述图1中的自车401中的智能驾驶装置，该方法可以包括以下步骤：
步骤501、智能驾驶装置获取当前时刻自车周围的环境信息、当前时刻自车的状态信息。
其中，环境信息表示该自车周围环境中一个或多个对象及各对象的属性信息，例如，自车道的车道线、道路的路沿边界、邻车道的车道线、目标车辆、前向车流、侧方车辆、后方车辆、邻车道中车辆、行人、交通信号灯、建筑、树木、施工区域等等对象，及各对 象的位置、速度、加速度、方位、距离等属性信息。示例性地，可以通过安装在自车车身四周的雷达(如，毫米波雷达、激光雷达、超声波雷达)、图像采集装置(如摄像头)等传感器感知自车周围的环境信息。
自车的状态信息可以包括：驾乘状态信息、运动状态信息等，其中，运动状态信息表示自车的运动学状态的信息，例如，自车的速度、加速度、横摆角速度、转向角度、转向半径、车身位置及姿态等信息。示例性地，可以通过安装于自车内部的传感器，实时获取运行状态信息。驾乘状态信息表示驾驶员设定的参数信息，例如，驾驶员设定的巡航速度，跟车距离，车头时距、驾驶员变道指令、变道时长等等。示例性地，可以接收驾驶员通过显示屏、方向盘的物理按键、语音等方式输入的驾乘状态信息。
需要说明的是，上述周围环境中各对象的速度、加速度、距离等可分别为各对象相对于自车的相对速度、相对加速度、相对距离；各对象的位置、方位等可以为车体坐标系中的位置或方位；其中，车体坐标系用来描述自车周围的对象和该自车之间的相对位置关系，可以通过预先标定明确各个传感器的坐标系与车体坐标系的之间的转换关系，从而根据传感器采集的数据，确定各对象在车体坐标系中的坐标。本申请实施例中，如无特殊说明，各对象的位置等均指在车体坐标系中的坐标；示例性地，以自车后(轮)轴中心在地面上投影为三维原点，沿车身向前为X轴的正方向，沿后轴轴线向左为Y轴的正方向，向上为Z轴的正方向，即形成了对应的车体坐标系XYZ。
步骤502、智能驾驶装置根据当前时刻自车周围的环境信息及自车的状态信息，确定当前时刻的控制点。
其中，控制点，又称虚拟点，可以为上述自车周围环境中的对象上的一个点，例如，可以为目标车辆后保险杠的中点，还可以为自车道的车道线上与自车距离一定长度的点，等等。控制点还可以为上述自车周围环境中的对象之外的一个点，例如，可以为自车道的左右两条车道线之间的一个点，还可以为自车正前方一定距离的点，等等。
下面对该步骤中确定当前时刻的控制点的实现方式进行示例性说明。方式一，可以根据当前时刻自车周围环境信息中所包含的第一对象及自车的状态信息(如驾乘状态信息)，确定当前时刻唯一的控制点。
在一种可能的实现方式中，可以根据环境信息中各第一对象的预设优先级信息及当前时刻的驾乘状态信息，依次确定当前时刻的控制点。在一示例中，可以根据各第一对象的预设优先级信息，确定在环境信息中存在的、优先级最高的第一对象，根据该优先级最高的第一对象，确定当前时刻的控制点。例如，可以根据各第一对象的预设优先级信息，依次判断各第一对象在当前时刻自车周围的环境信息中是否存在；当确定一第一对象在当前时刻自车周围的环境信息中存在时，根据存在的该第一对象，确定当前时刻的控制点。
其中，第一对象可以包括自车行驶的参照对象，例如，可以为自车道的车道线、目标车辆、邻车道中车辆等等；不同的第一对象对应于不同的特性，各第一对象对应的特性可以为OBF、LKA、TJA、LCC等特性。即预设优先级也可表征ADAS系统中预设的处理各特性的优先级排序。示例性地，自车道的车道线、目标对象、邻车道的车道线的预设优先级依次降低。这样，根据环境信息中各第一对象的预设优先级信息及驾乘状态信息，依次确定当前时刻的控制点，这样，通过多个信息源(即环境信息中各第一对象)，按照各信息源的优先级排序的方式，改变不同场景的控制点的位置，确定当前时刻的控制点，从而实现 多个特性(如：OBF、LKA、TJA、LCC等特性)，同时，根据多个信息源的优先级，实现不同特性之间的切换，保证了多个特性或场景间的平滑过渡。无需利用状态机实现不同特性之间的切换，降低了逻辑复杂度。
需要说明的是，在不同的ADAS解决方案中，可以包括一个或多个特性，不同的特性对应不同的第一对象，其中，特性的数量、第一对象的类型及各类型的预设优先级可以由ADAS系统预先配置，或者由驾驶员进行设定。
示例性地，第一对象可以包括：自车道的车道线、自车所在车道中最接近的车辆(closest in-path vehicle，CIPV)(即目标车辆)、邻车道中车辆；其中，自车道的车道线、CIPV、邻车道中车辆的优先级依次降低。在自车道的车道线存在的情况下，将自车道的车道线上预设距离对应的、且符合预设条件的位置点作为当前时刻的控制点；如果自车道的车道线不存在，则在CIPV存在的情况下，将该CIPV对应的、且符合预设条件的位置点作为当前时刻的控制点；如果CIPV也不存在，则在邻车道中车辆存在的情况下，将该邻车道中车辆在自车道投影的、且符合预设条件的位置点作为当前时刻的控制点。
其中，自车道的车道线存在、CIPV存在、邻车道中车辆存在等表示自车的传感器可以感知到自车道的车道线、CIPV及邻车道中车辆等的数据，同时，该数据满足预设的对数据的稳定性及准确性的要求。预设条件可以包括：位置点在自车当前行驶状态下可到达的范围之内，例如，自车当前为向前行驶状态，预设条件为位置点的X轴坐标值大于3m，即若位置点位于自车的前方且X轴坐标值大于3m时，则符合预设条件，若位置点位于自车的后方或侧方，则不符合预设条件。预设距离可以基于驾乘状态信息确定，例如，驾驶员设定的跟车距离为15～40m，则预设距离可以包括：位置点的X轴坐标值为15～40m等等，即取自车道的车道线上与车辆的纵向距离相差15～40m的位置点。自车道的车道线上预设距离对应的位置点可以为自车道的车道线上预设距离上的点，可以为自车道的车道线上预设距离上的点投影到自车道的车道中心线上的点，等等；CIPV对应的位置点可以为CIPV后保险杠中点，还可以为CIPV投影到自车道的车道中心线上的点，等等。邻车道中车辆在自车道的投影的位置点可以为邻车道中车辆后保险杠中点横向投影到自车道的投影点，例如，将自车道左边的邻车道中车辆后保险杠中点在横向上向右移动一定距离(如一个车道宽度)，所对应的投影点；其中，横向为与车道线所指示的行驶方向垂直的方向，相应的，纵向为车道线所指示的行驶方向。
例如，图3示出根据本申请一实施例的一种确定当前时刻控制点的流程图，如图3所示，智能驾驶装置首先判断环境信息中是否存在自车道的车道线，在自车道的车道线存在的情况下，获取自车道的车道线上预设距离对应的位置点，若该位置点满足预设条件，则将该位置点作为当前时刻的控制点，若该位置点不满足预设条件，则退出ADAS系统，并提醒驾驶员接管车辆控制；在自车道的车道线不存在的情况下，则进一步判断环境信息中是否存在CIPV。在CIPV存在的情况下，获取CIPV对应的位置点，若该位置点满足预设条件，则将该位置点作为当前时刻的控制点，若该位置点不满足预设条件，则退出ADAS系统，并提醒驾驶员接管车辆控制；在CIPV不存在的情况下，则进一步判断邻车道中车辆是否存在。在邻车道中车辆存在的情况下，获取邻车道中车辆在自车道投影的位置点，若该位置点满足预设条件，则将该位置点作为当前时刻的控制点，若该位置点不满足预设条件，则退出ADAS系统，并提醒驾驶员接管车辆控制。在邻车道中车辆不存在的情况下，则退 出ADAS系统，并提醒驾驶员接管车辆控制。
其中，在自车道的车道线存在且确定了当前时刻的控制点时，进而生成控制轨迹，可以实现LCC、LKA等特性，在CIPV存在时且确定了当前时刻的控制点时，进而生成控制轨迹，可以实现TJA、OBF等特性，在邻车道中车辆存在且确定了当前时刻的控制点时，进而生成控制轨迹，可以实现TJA、LKA等特性。这样，利用自车道的车道线、目标车辆、邻车道中车辆，按照优先级排序的方式，改变不同场景的控制点的位置，可以根据环境信息中存在的、优先级最高的第一对象确定当前时刻的控制点，从而实现多个特性，例如，可以根据当前时刻自车周围的环境信息中自车道的车道线确定当前时刻的控制点，进而实现LCC、LKA等特性；可以根据当前时刻自车周围的环境信息中CIPV，确定当前时刻的控制点，进而实现TJA、OBF等特性；可以根据当前时刻自车周围的环境信息中自车的邻车道中车辆，确定当前时刻的控制点，进而实现TJA、LKA等特性；同时，根据自车道的车道线、目标车辆、邻车道中车辆的优先级，实现不同特性之间的切换，保证了多个特性或场景间的平滑过渡。
方式二，可以根据当前时刻自车周围的环境信息中所包含的第一对象、第二对象及自车的状态信息(如驾乘状态信息)，确定当前时刻唯一的控制点。
在一种可能的实现方式中，根据环境信息中各第一对象的预设优先级信息及当前时刻的驾乘状态信息，依次确定所述当前时刻的控制点，可以包括：获取优先级最高的第一对象对应的位置点；判断预设区域内是否存在第二对象；在第二对象存在时，对第一对象对应的位置点进行偏移处理，并将偏移后的位置点作为当前时刻的控制点。
其中，第一对象对应的位置点可以为上述自车道的车道线上预设距离对应的位置点、CIPV对应的位置点、邻车道中车辆在自车道投影的位置点等等，第二对象为环境信息中自车的避让对象，例如，可以为侧方车辆、后方车辆、行人、施工区域、道路的路沿边界等等，不同的第二对象对应不同的特性，各第二对象对应的特性可以为CAA、路沿远离等特性。预设区域为道路中具有一定大小的区域，例如，可以为以第一对象对应的位置点为中心的具有一定大小的圆形区域，还可以为自车道中具有一定大小的矩形区域。在上述预设区域中存在第二对象时，说明若以上述获取的第一对象对应的位置点生成控制轨迹，并按照该控制轨迹控制自车行驶，存在碰撞第二对象的风险，此时，对第一对象对应的位置点进行偏移处理，例如，横向偏移处理、纵向偏移处理，并将偏移后的位置点作为当前时刻的控制点，从而避免碰撞到第二对象，这样，基于第一对象，按照优先级排序的方式，并结合存在的第二对象，共同改变不同场景的控制点的位置，确定当前时刻的控制点，可以实现多个特性(如CAA、路沿远离等特性)，并保证了多个特性或场景间的平滑过渡。
需要说明的是，在不同的ADAS解决方案中，可以包括一个或多个特性，不同的特性对应不同的第二对象，其中，特性的数量、第二对象的类型及预设区域的大小可以由ADAS系统预先配置，或者由驾驶员进行设定。
示例性地，第二对象包括车道的硬边界；预设区域为以第一对象对应的位置点为中心预设大小的第一区域。
在一种可能的实现方式中，上述在第二对象存在时，对第一对象对应的位置点进行偏移处理，并将偏移后的位置点作为当前时刻的控制点，包括：在第一区域内存在车道的硬边界时，对第一对象对应的位置点进行偏移处理，以使硬边界在以偏移后的位置点为中心 预设大小的区域之外，且偏移后的位置点在自车道内；将偏移后的位置点作为当前时刻的控制点。
其中，硬边界可以是自车不能触碰或跨越的边界，例如可以包括：自车道的实线车道线、道路的路沿边界等，第一区域可以为以第一对象对应的位置点为中心的具有一定大小的区域，例如，可以为以第一对象对应的位置点为圆心，半径为0.5～1.875米，所构成的圆形区域。偏移处理可以为基于自车道的车道中心线的位置，将第一对象对应的位置点在横向上移动一定距离，例如，自车道的车道中心线位于第一对象对应的位置点的左边，则将第一对象对应的位置点在横向上向左移动一定距离，以使该硬边界在以偏移后的位置点为中心的具有一定大小的区域之外，且该偏移后的位置点在自车道内；将偏移后的位置点作为当前时刻的控制点，该控制点远离车道的硬边界，从而使得自车根据该控制点生成的控制轨迹行驶时，避免碰撞到硬边界，这样，基于第一对象，按照优先级排序的方式，并结合硬边界，共同改变不同场景的控制点的位置，确定当前时刻的控制点，可以实现多个特性(如路沿远离等特性)，并保证了多个特性或场景间的平滑过渡。
示例性地，第二对象可以包括：侧方车辆、行人等障碍物；预设区域可以为自车道中自车前方预设大小的第二区域。
在一种可能的实现方式中，上述在第二对象存在时，对第一对象对应的位置点进行偏移处理，并将偏移后的位置点作为当前时刻的控制点，包括：在第二区域内存在障碍物时，根据障碍物对应的权重信息，对第一对象对应的位置点进行偏移处理，以使该障碍物在以偏移后的位置点为中心预设大小的区域之外，且该偏移后的位置点在自车道内；将偏移后的位置点作为当前时刻的控制点。
其中，第二区域可以为自车道中自车前方具有一定纵向长度(如10～40m)及一定横向宽度(如1.875～3.75m)的矩形区域。障碍物对应的权重信息表示自车避让该障碍物的程度，例如，权重信息的数值越大，则自车避让该障碍物的动作越明显，相应的偏移后的位置点与第一对象对应的位置点之间的距离越远；权重信息的数值越小，则自车避让该障碍物的动作越小，相应的偏移后的位置点与第一对象对应的位置点之间的距离越近。这样，考虑到不同的场景中，障碍物的纵向距离、横向距离、相对速度等信息不同，自车避让障碍物的动作不同，利用障碍物对应的权重信息表示不同场景下自车避让该障碍物的程度，并根据障碍物对应的权重信息，对第一对象对应的位置点进行偏移处理，从而实现不同场景下的位置点偏移，将偏移后的位置点作为当前时刻的控制点，该控制点远离障碍物，从而使得自车根据该控制点生成的控制轨迹行驶时，避免碰撞到该障碍物，这样，基于第一对象按照优先级排序的方式，并结合障碍物及障碍物对应的权重信息，共同改变不同场景的控制点的位置，确定当前时刻的控制点，从而实现多个特征(如CAA等特性)，并保证了多个特性或场景间的平滑过渡。
示例性地，权重信息的数值可以由预设的权重函数(或权重体函数)确定，该权重函数所包含的参数可以包括：车头时距(Time of Headway，THW)、碰撞时间(Time to Collision，TTC)、侵入宽度(Invade Vehicle Width，IVW)等等。其中，THW可以由驾乘状态信息确定，即可以由驾驶员预先设定，TTC可以根据环境信息中入侵车辆与自车的相对距离及相对速度确定，IVW可以根据环境信息中自车道的车道线的位置及入侵车辆的位置确定。以权重函数所包含的参数为THW、TTC、IVW为例，权重信息的数值可以表示为α＝ f(THW,TTC,IVW)，其中，α为权重信息的数值，f表示权重体函数，权重信息的数值α与THW成负相关，即THW越小，自车避让障碍物的动作越明显，权重信息的数值α与TTC成负相关，即TTC越小，自车避让障碍物的动作越明显，权重信息的数值α与IVW成正相关，即IVW越大，自车避让障碍物的动作越明显；这样，根据不同场景下对应的权重信息的数值，改变不同场景下控制点的位置，确定当前时刻的控制点，实现了不同场景的平滑过渡。
下面以障碍物为侵入车辆为例，对不同场景下，障碍物对应的权重信息进行说明。
图4示出根据本申请一实施例的多种车辆入侵自车道场景的示意图，在图4(a)-图4(f)中，箭头的方向表示车辆的速度方向，其中，自车401及入侵车辆701的车头前方的箭头所指的方向为正向，入侵车辆701车尾上的箭头所指的方向为反向，箭头的数量表示车辆的速度大小，箭头的数量越多，则车辆的速度越大，其中，图4(a)表示入侵车辆近距离保持的场景，即入侵车辆701与自车401的速度方向均为正向，且大小近似相等；图4(b)表示入侵车辆高速略过的场景，即入侵车辆701与自车401的速度方向均为正向，且入侵车辆701的速度大小大于自车401的速度大小；图4(c)表示入侵车辆低速逼近的场景，即入侵车辆701的速度方向为反向，自车401的速度方向为正向，且入侵车辆701的速度大小小于自车401的速度大小；图4(d)表示入侵车辆等速切入的场景，即入侵车辆701与自车401的速度方向均为正向，且成一定夹角，且入侵车辆701的速度等于自车401的速度；4(e)表示入侵车辆快速切入的场景，即入侵车辆701与自车401的速度方向均为正向，且成一定夹角，且入侵车辆701的速度大于自车401的速度；4(f)表示入侵车辆低速切入的场景，即入侵车辆701的速度方向为反向，自车401的速度方向为正向，且两者成一定夹角，且入侵车辆701的速度小于自车401的速度。
图5示出根据本申请一实施例的针对图4中多种车辆入侵自车道场景中位置点偏移的示意图，如图5所示，“0”点表示自车道的车道中心线上预设距离的点，“1”点表示入侵车辆的后保险杠在远离入侵车辆的自车道的一条车道线上的横向投影点。设“0”点为第一对象对应的位置点，则偏移后的位置点为以“0”点与“1”点为端点的线段上的点，若以“0”点作为偏移后的位置点，即以“0”点作为控制点生成控制轨迹，引导自车行驶，则自车不会产生避让动作，若以“1”点作为偏移后的位置点，即以“1”点作为控制点生成控制轨迹，引导自车行驶，则自车会有明显的避让动作，若以“0”点与“1”点为端点的线段上的点作为偏移后的位置点，则自车会产生避让动作，避让的程度(或避让动作的幅度)与偏移后的位置点相对于1点的距离负相关，即偏移后的位置点距离1点越近，避让动作幅度越大。针对上述图4(a)-图4(f)所示的不同的车道入侵场景，若以图5中“0”点作为偏移后的位置点，则自车不会产生避让动作，可以适用于图4(b)中入侵车辆高速略过的场景；若以“1”点作为偏移后的位置点，则自车会有明显的避让动作，可以适用于图4(f)表示入侵车辆低速切入的场景。若以“0”点与“1”点之间的点作为偏移后的位置点，则该点距离“0”点越近，自车避让动作越不明显，可以适用于图4(a)中入侵车辆近距离保持的场景及图4(c)中入侵车辆低速逼近的场景；该点距离“1”点越近，自车避让动作越明显，可以适用于图4(d)中入侵车辆等速切入的场景及图4(e)中入侵车辆快速切入的场景。
不同的场景中，可以根据环境信息中侵入车辆的位置、相对距离、相对速度等，基于上述权重体函数f，确定侵入车辆对应的权重信息的数值，即α＝f(THW,TTC,IVW)，该权 重信息的不同取值，即对应了在“0”，“1”连线上的不同点，例如，权重信息的取值在0～1之间，将“0”与“1”之间的线段按比例等分，当权重信息的数值等于1时，对应“1”点为偏移后的位置点；当权重信息的数值等于“0”时，对应“0”点为偏移后的位置点；当权重信息的数值大于0且小于1时，则选取该线段中对应位置的点。这样，在不同场景下，可以根据THW、TTC、IVW及权重体函数f，确定偏移后的位置点，若以该偏移后的位置点作为当前时刻的控制点生成当前时刻的控制轨迹，可以保证自车沿该控制轨迹行驶不会与入侵车辆碰撞。
例如，上述图4中侵入车辆701与自车401在不同场景中，相对位置相同，相对速度不同时，TTC不同，此时所对应的权重体函数形状不同。图6示出根据本申请一实施例的一种权重体函数的示意图，如图6(a)-6(d)所示，TTC＝-4时，权重信息的数值α关于THW、IVW的权重体函数，其中，6(a)-6(c)为图6(d)的三视图。随着TTC的变化，权重体函数的形状也会随之改变，对应于不同场景下的避让障碍物情形。图7示出根据本申请一实施例的另一种权重体函数的示意图，如图7(a)-7(d)所示，不同TTC时对应的权重体函数变化趋势；其中，图7(a)示出TTC<0时，α关于THW、IVW的权重体函数；图7(b)示出TTC＝∞时，α关于THW、IVW的权重体函数；图7(c)示出TTC＝20时，α关于THW、IVW的权重体函数；图7(b)示出TTC＝4时，α关于THW、IVW的权重体函数。
图8示出根据本申请一实施例的一种确定当前时刻控制点的流程图，如图8所示，智能驾驶装置首先判断环境信息中是否存在自车道的车道线，在自车道的车道线存在的情况下，获取自车道的车道线上预设距离对应的位置点；在自车道的车道线不存在的情况下，则进一步判断环境信息中是否存在CIPV。在CIPV存在的情况下，获取CIPV对应的位置点；在CIPV不存在的情况下，则进一步判断邻车道中车辆是否存在。在邻车道中车辆存在的情况下，获取邻车道中车辆在自车道投影的位置点；在邻车道中车辆不存在的情况下，则退出ADAS系统，并提醒驾驶员接管车辆控制。
进一步地，在获取自车道的车道线上预设距离对应的位置点、CIPV对应的位置点、邻车道中车辆在自车道投影的位置点中任一位置点后，进而判断第一区域内是否存在车道的硬边界及第二区域内是否存在障碍物，其中，执行判断第一区域内是否存在车道的硬边界与判断第二区域内是否存在障碍物的步骤不分先后。在第一区域内存在车道的硬边界或第二区域内存在障碍物时，进行相应的位置点偏移处理，得到偏移后的位置点，若偏移后的位置点满足预设条件，则将该偏移后的位置点作为当前时刻的控制点，若偏移后的位置点不满足预设条件，则退出ADAS系统，并提醒驾驶员接管车辆控制。在第一区域内不存在车道的硬边界且第二区域内不存在障碍物时，则判断该位置点是否满足预设条件，若该位置点满足预设条件，则将该位置点作为当前时刻的控制点，若该位置点不满足预设条件，则退出ADAS系统，并提醒驾驶员接管车辆控制。
其中，在自车道的车道线存在且无需进行位置点偏移处理时，进而可以根据确定的控制点生成控制轨迹，可以实现LCC、LKA等特性，在CIPV存在且无需进行位置点偏移处理时，进而可以根据确定的控制点生成控制轨迹，可以实现TJA、OBF等特性，在邻车道中车辆存在且无需进行位置点偏移处理时，进而可以根据确定的控制点生成控制轨迹，可以实现TJA、LKA等特性；在自车道的车道线、CIPV、邻车道中车辆中任一个存在，且第一区 域内存在车道的硬边界时，进而可以根据确定的控制点生成控制轨迹，可以实现路沿远离等特性。在自车道的车道线、CIPV、邻车道中车辆中任一个存在，且第二区域内存在障碍物时，进而可以根据确定的控制点生成控制轨迹，可以实现CAA等特性。这样，可以根据当前时刻自车周围的环境信息中的第一对象及第一对象预设的优先级信息，第二对象及第二对象对应的权重信息等，实现不同的特性，并保证多个特性及场景间的平滑过渡。
示例性地，图9示出根据本申请一实施例的一种利用控制点实现OBF特性的示意图，OBF特性要求在无车道线、只有在自车前方的车辆时，自车能够跟随前方车辆的轨迹行驶。如图9所示，当前时刻环境信息中自车道的车道线不存在，CIPV1101存在，此时选取CIPV后保险杠中点1102为位置点，同时，第一区域内不存在车道的硬边界，第二区域内不存在障碍物，且该位置点满足预设条件，则将该位置点，即CIPV后保险杠中点1102作为控制点，进而生成当前时刻的控制轨迹，实现自车401的OBF特性。
示例性地，图10示出根据本申请一实施例的一种利用控制点实现LCC特性的示意图，如图10所示，当前时刻环境信息中自车道的车道线存在，此时选取自车道的两条车道线上纵向上距离自车15～40m的点，从而得到自车道的车道中心线上纵向上距离自车15～40m的点1202，将该点作为位置点，同时，第一区域内不存在车道的硬边界，第二区域内不存在障碍物，且该位置点满足预设条件，则将该位置点作为控制点，进而生成当前时刻的控制轨迹，实现自车401的LCC特性。
示例性地，图11示出根据本申请一实施例的一种利用控制点实现TJA特性的示意图，TJA特性要求在有车道线，但由于前车遮挡而导致自车的传感器无法准确感知到车道线的位置与形状信息，即传感器感知到的车道线数据无法满足LCC特性对稳定性和准确性要求的情况下，借助前车进行横向控制；同时，TJA不能控制自车完全根据前车轨迹行驶，当前车在自车道内偏移(如图11(a)所示)或有切入切出动作(如图11(b)所示)时，仍然需要保持自车在自车道内。如图11(a)-11(b)所示，当前环境中自车道的右车道线的一部分被CIPV遮挡，CIPV1301存在，此时选取CIPV后保险杠中点1302，将自车道的车道中心线作为参考线，将点1302投影到自车道的车道中心线上，即点1303，将该点1303作为位置点，同时，第一区域内不存在车道的硬边界，第二区域内不存在障碍物，且该位置点满足预设条件，则将该位置点1303作为控制点，进而生成当前时刻的控制轨迹，从而保证CIPV在自车道内偏移或有切入切出动作时，自车依然能够稳定保持居中，实现自车401的TJA特性。
示例性地，图12示出根据本申请一实施例的一种利用控制点实现ALC/SLC特性的示意图，实现ALC(如图12(a)-12(b)所示)及SLC(如图12(c)所示)特性的过程实质是将自车道中的控制点逐渐平移到临车道的车道中心线的过程。如图12(a)所示，响应于驾驶员的变道指令，在变道的过程中，判断当前时刻环境信息中自车道的车道线存在，此时选取自车道的车道线上纵向上距离自车401预设距离的点，并将该点在横向上向邻车道偏移一定距离(该距离可以根据预设的变道时长、自车道中心线与邻车道中心线的之间的间距等确定)，从而得到位置点1401，同时，第一区域内不存在车道的硬边界，第二区域内不存在障碍物，且该位置点满足预设条件，则将该位置点作为控制点，进而生成当前时刻的控制轨迹，实现自车401的SLC特性。如图12(b)所示，响应于驾驶员的变道指令，在变道的过程中，判断当前时刻环境信息中自车道的车道线存在，此时选取自车道的 车道线上纵向上距离自车401预设距离的点，并将该点在横向上向邻车道偏移一定距离，得到位置点，同时，第二区域内存在障碍物1402，则进行位置点偏移处理，将偏移后且满足预设条件的位置点1403作为控制点，进而生成当前时刻的控制轨迹，实现自车401的SLC避让特性。如图12(c)所示，在前车1404阻挡自车道的通行路线时，进行变道，在变道的过程中，判断当前时刻环境信息中自车道的车道线存在，此时选取自车道的车道线上纵向上距离自车401预设距离的点，并将该点在横向上向邻车道偏移一定距离，得到位置点，同时，第二区域内存在障碍物1405及1404，则进行位置点偏移处理，将偏移后且满足预设条件的位置点1406作为控制点，进而生成当前时刻的控制轨迹，实现自车401的ALC特性。
步骤503、根据当前时刻的控制点、历史时刻的控制轨迹及当前时刻自车的状态信息(如运动状态信息)，生成当前时刻的控制轨迹，控制轨迹表示用于导引自车行驶的轨迹。
其中，历史时刻的控制轨迹可以包括上一时刻的控制点生成的控制轨迹，可以理解的是，随着实时生成控制轨迹，可以获取历史时刻的控制轨迹；历史时刻的控制轨迹可以表征时间域上的不同时刻的控制点的位置，这样，利用空间上的控制点在时间域上的连续性生成控制轨迹，即只需要空间上一个点即可在时间维度生成控制曲线，将环境信息中各种对象降维到一个点上，相比于在空间域上取多个点生成控制轨迹的方式，本申请实施例无需预先生成多个轨迹线，大大降低了对算力的要求，有效提高了ADAS系统性能。
例如，图13示出根据本申请一实施例的一种智能驾驶的对比图，如图13所示，目标车辆1505真实的轨迹为1501；1502为通过该真实的轨迹1501，在空间上用最小二乘法拟合生成的OBF特性的控制轨迹；1503为通过t1时刻的控制点，t1时刻之前的控制轨迹及t1时刻运动状态信息，生成的t1时刻的OBF特性的控制轨迹，1104为通过t2时刻的控制点，t2时刻之前的控制轨迹(包括t1时刻的控制轨迹)及t2时刻运动状态信息，生成的t2时刻的OBF特性的控制轨迹。轨迹1501为通过轨迹1501中的多个点在空间维度拟合生成，而轨迹1503及1504为通过目标车辆1505确定各时刻的控制点，基于不同时刻的控制点在时间维度拟合生成。
示例性地，控制轨迹可以为三阶曲线，这样，利用当前时刻的控制点、历史时刻的控制轨迹及当前时刻运动状态信息，生成一条三阶曲线，该三阶曲线可以与一些ADAS系统架构适配；从而提高了本申请中智能驾驶方法的适用范围。
在一种可能的实现方式中，可以根据当前时刻的控制点、历史时刻的控制轨迹及当前时刻的自车状态信息，利用滤波器，生成当前时刻的控制轨迹。示例性地，该滤波器可以为卡尔曼滤波器(kalman filte)，卡尔曼滤波器具有时间连续性，将当前时刻控制点的位置信息、历史时刻的控制轨迹及当前时刻自车状态信息，输入到卡尔曼滤波器，利用空间上的控制点在时间域上的连续性生成当前时刻的控制轨迹，从而实现不同的特性，提高了系统性能。可以理解的是，随着时间的推移，不同时刻的控制轨迹不断生成，历史时刻的控制轨迹随之不断迭代。
示例性地，可以将上述图10中CIPV后保险杠中点1102的位置信息、历史时刻的控制轨迹及当前时刻自车的速度，横摆角速度、转向半径等信息输入到卡尔曼滤波器生成当前时刻的控制轨迹，并利用该控制轨迹进行循迹控车，从而实现OBF特性。
示例性地，可以将上述图11中自车道的车道中心线上纵向上距离自车20m的点1202 的位置信息、历史时刻的控制轨迹及当前时刻自车的速度，横摆角速度、转向半径等信息输入到卡尔曼滤波器，生成当前时刻的控制轨迹，并利用该控制轨迹进行车道居中控制，从而实现LCC特性。
示例性地，可以将上述图12中CIPV后保险杠的中点1302投影到自车道的车道中心线上的点1303的位置信息、历史时刻的控制轨迹及当前时刻自车的速度，横摆角速度、转向半径等信息输入到卡尔曼滤波器，生成当前时刻的控制轨迹，从而实现TJA特性。同时，利用车道中心线上的点生成控制轨迹而不是直接用车道中心线作为控制轨迹，保证了TJA特性中控制的平顺性。
示例性地，可以将图13中变道过程中的当前时刻控制点的位置信息、历史时刻的控制轨迹及当前时刻自车的速度，横摆角速度、转向半径等信息输入到卡尔曼滤波器，生成当前时刻的控制轨迹，并利用该控制轨迹进行变道控制，从而实现ALC/SLC特性。进一步，在实现ALC/SLC特性的过程中，为了保证变道过程符合驾驶员习惯，通过对大量驾驶员变道数据进行分析，预先得到变道时不同变道时长对应的横向偏移速度曲线，该横向偏移速度曲线表示自车变道时的各时刻的横向速度变化，图14示出根据本申请一实施例的一种横向偏移速度曲线的示意图，如图14所示，驾驶员变道时的横向速度变化符合V形曲线，针对不同变道时长，会在V形的基础上下凹或上突。在变道过程中，可以根据预设或预估的变道时长，选取对应的横向偏移速度曲线，根据该横向偏移速度曲线确定各时刻的对应的横向速度，进而生成各时刻的控制轨迹，达到舒适的变道效果。
需要说明的是，上述利用控制点实现的ADAS系统的几种特性仅为示例，可以理解的是，本申请实施例的利用控制点实现的特性不限于此，对于ADAS系统的中涉及横向控制及纵向控制的其他特性均可适用；还可适用于未来演进的特性，例如，多目标协同(Multi-object Coordination，MOC)，可以根据目标车辆、临车道车辆、前前车辆(即位于自车的前方，且与自车相隔至少一个车辆的车辆)等环境信息，一同作用确定控制点，进而生成控制轨迹，从而改善临车道车辆近距离切入、临车道车辆减速切入、前前车辆刹车且前车切出等场景下ACC特性的体验。
本申请实施例中，通过获取当前时刻自车周围的环境信息及当前时刻自车状态信息；根据环境信息及状态信息，确定当前时刻的控制点；根据控制点、历史时刻的控制轨迹及状态信息，生成当前时刻的控制轨迹，控制轨迹表示用于导引自车行驶的轨迹。这样，基于驾驶员驾驶习惯及思维方式，当前时刻驾驶员通常将目光集中在空间中的一点，该点即为驾驶员期望车辆能够行驶到的位置，将该点作为控制点，将环境信息对特性的影响集中在该控制点，通过该控制点生成当前时刻合理的控制轨迹，满足不同场景下的控制要求，且无需预先对环境信息进行处理生成多个条轨迹线，大大降低了对算力的要求，从而提升了ADAS系统的性能。还可以根据环境信息中各第一对象的预设优先级信息及状态信息，依次确定当前时刻的控制点，这样，通过多个信息源(即环境信息中各第一对象)，按照各信息源的优先级排序的方式，改变不同场景下控制点的位置，确定当前时刻的控制点，从而实现多个特性(如：OBF、LKA、TJA、LCC等特性)，同时，根据多个信息源的优先级及对应的权重信息，实现不同特性之间的切换，保证了多个特性或场景间的平滑过渡，且无需利用状态机实现不同特性之间的切换，降低了逻辑复杂度。
基于上述方法实施例的同一发明构思，本申请的实施例还提供了一种智能驾驶装置， 该智能驾驶装置用于执行上述方法实施例所描述的技术方案。
图15示出根据本申请一实施例的一种智能驾驶装置的结构示意图；如图15所示，该智能驾驶装置可以包括：获取模块1701，用于获取当前时刻自车周围的环境信息及当前时刻自车状态信息；确定模块1702，用于根据所述环境信息及状态信息，确定当前时刻的控制点；生成模块1703，用于根据所述控制点、历史时刻的控制轨迹及所述状态信息，生成当前时刻的控制轨迹，所述控制轨迹表示用于导引自车行驶的轨迹。
在一种可能的实现方式中，所述确定模块1702，还用于：根据所述环境信息中各第一对象的预设优先级信息及所述状态信息，依次确定所述当前时刻的控制点。
在一种可能的实现方式中，所述第一对象至少包括所述自车的自车道的车道线、目标车辆、所述自车的邻车道中车辆的一个；其中，所述自车道的车道线、所述目标车辆、所述邻车道中车辆的优先级依次降低；
所述确定模块1702，还用于：在所述自车道的车道线存在的情况下，将所述自车道的车道线上预设距离对应的、且符合预设条件的位置点作为当前时刻的控制点；所述预设距离基于所述状态信息确定；如果所述自车道的车道线不存在，则在所述目标车辆存在的情况下，将所述目标车辆对应的、且符合预设条件的位置点作为当前时刻的控制点；如果所述自车道的车道线不存在，且所述目标车辆不存在，则在所述邻车道中车辆存在的情况下，将所述邻车道中车辆在自车道投影的、且符合预设条件的位置点作为当前时刻的控制点。
在一种可能的实现方式中，所述确定模块1702，还用于：获取优先级最高的第一对象对应的位置点；判断预设区域内是否存在第二对象，所述第二对象为环境信息中自车的避让对象；在所述第二对象存在时，对所述第一对象对应的位置点进行偏移处理，并将偏移后的位置点作为当前时刻的控制点。
在一种可能的实现方式中，所述第二对象包括自车道的硬边界；所述预设区域为以第一对象对应的位置点为中心预设大小的区域；
所述确定模块1702，还用于：在所述自车道的硬边界存在时，对所述第一对象对应的位置点进行偏移处理，以使所述硬边界在以所述偏移后的位置点为中心预设大小的区域之外，且所述偏移后的位置点在自车道内；将偏移后的位置点作为当前时刻的控制点。
在一种可能的实现方式中，所述第二对象包括障碍物；所述预设区域为所述自车道中自车前方预设大小的区域；
所述确定模块1702，还用于：在所述障碍物存在时，根据所述障碍物对应的权重信息，对所述第一对象对应的位置点进行偏移处理，以使所述障碍物在以所述偏移后的位置点为中心预设大小的区域之外，且所述偏移后的位置点在自车道内；其中，所述权重信息表示自车避让该障碍物的程度；将偏移后的位置点作为当前时刻的控制点。
在一种可能的实现方式中，所述权重信息的数值至少由车头时距THW、碰撞时间TTC、侵入宽度IVW中的一个确定，其中，所述权重信息的数值与所述THW成负相关，与所述TTC成负相关，与所述IVW成正相关。
在一种可能的实现方式中，所述生成模块1703，还用于：根据所述控制点、历史时刻的控制轨迹及状态信息，利用卡尔曼滤波器，生成所述当前时刻的控制轨迹。
本申请实施例中，通过获取当前时刻自车周围的环境信息及当前时刻自车状态信息； 根据环境信息及状态信息，确定当前时刻的控制点；根据控制点、历史时刻的控制轨迹及状态信息，生成当前时刻的控制轨迹，控制轨迹表示用于导引自车行驶的轨迹。这样，基于驾驶员驾驶习惯及思维方式，当前时刻驾驶员通常将目光集中在空间中的一点，该点即为驾驶员期望车辆能够行驶到的位置，将该点作为控制点，将环境信息对特性的影响集中在该控制点，通过该控制点生成当前时刻合理的控制轨迹，满足不同场景下的控制要求，且无需预先对环境信息进行处理生成多个条轨迹线，大大降低了对算力的要求，从而提升了ADAS系统的性能。还可以根据环境信息中各第一对象的预设优先级信息及状态信息，依次确定当前时刻的控制点，这样，通过多个信息源(即环境信息中各第一对象)，按照各信息源的优先级排序的方式，改变控制点的位置，确定当前时刻的控制点，不同的第一对象对应于不同的特性，从而实现多个特性(如：OBF、LKA、TJA、LCC等特性)，同时，根据多个信息源的优先级及对应的权重信息，实现不同特性之间的切换，保证了多个特性或场景间的平滑过渡，且无需利用状态机实现不同特性之间的切换，降低了逻辑复杂度。
上述实施例的各种可能的实现方式或说明参见上文，此处不再赘述。
本申请实施例还提供了一种新的ADAS系统构架，图16示出根据本申请一实施例的ADAS系统架构图。如图16所示，该ADAS系统架构包括：预处理层、规划层、决策层、控制层等，还可以包括：执行层等等。
其中，预处理层：用于生成当前时刻自车周围的环境信息；示例性地，可以对安装在自车上的传感器所检测到的数据进行处理，得到自车周围的环境信息。另外，还可以基于GPS数据、高精地图数据等，得到地图信息，例如，车辆与前方路口的距离，车辆所在的车道等等。
规划层：用于根据所述环境信息及当前时刻自车的状态信息，确定当前时刻的控制点；根据所述控制点、历史时刻的控制轨迹及所述状态信息，生成当前时刻的控制轨迹；所述控制轨迹表示用于导引自车行驶的轨迹；示例性地，该层以预处理层得到的环境信息作为输入，结合自车的状态信息(可以包括：驾乘状态信息及自车状态信息)，通过智能驾驶装置，执行本申请的智能驾驶方法，从而确定当前场景适用的特性(或称为功能)及该特性对应的控制轨迹，将该控制轨迹输入到决策层。
进一步地，规划层还可以根据所述环境信息中各第一对象的预设优先级信息及所述状态信息，依次确定所述当前时刻的控制点。还可以获取优先级最高的第一对象对应的位置点；判断预设区域内是否存在第二对象，所述第二对象为环境信息中自车的避让对象；在所述第二对象存在时，对所述第一对象对应的位置点进行偏移处理，并将偏移后的位置点作为当前时刻的控制点。
决策层：用于对所述控制轨迹是否适用于当前时刻自车的工作状态进行判断；示例性地，该层以规划层生成的控制轨迹作为输入，根据当前时刻自车中各部件(如传感器、发动机、电动助力转向等系统)的工作状态，对该控制轨迹能够适用于当前自车的工作状态进行判断，当判断该控制轨迹能适用当前时刻自车的工作状态时，将该控制轨迹输入到控制层。
控制层：用于在所述控制轨迹适用于当前时刻自车的工作状态的情况下，基于所述控制轨迹，生成当前时刻的控制信号，所述控制信号用于车辆的辅助驾驶；示例性地，该层以决策层的控制轨迹为输入，结合车辆的动力学信息，生成方向盘转角、加速度等控制信 号。
进一步地，该系统还可以包括执行层：以控制层生成的控制信号作为输入，执行该控制信号，例如，通过电动助力转向系统(Electric Power Steering，EPS)及发动机执行该控制信号，从而实现车辆的辅助驾驶。
在上述ADAS系统中，通过规划层，将环境信息中的各对象对特性的影响集中在空间的点上，将该点作为控制点，通过该控制点在时间维度上的连续性，生成控制轨迹，从而实现不同的特性。这样，仅需用于通过控制点生成控制轨迹的轨迹生成器，无需多个轨迹生成器，大大降低了对算力的要求，提高了系统性能。同时，相对于通过状态机，在复杂的切换条件与过渡逻辑下实现场景判断与切换的方案，在上述ADAS系统中，规划层通过环境信息中不同对象预设的优先级与对应的权重信息确定每一时刻控制点，实现不同特性及场景平滑过渡。另外，在上述ADAS系统中，若有新增特性，仅需在预处理层增加针对该特性对应的对象的处理，还可以设定新增特性对应对象的优先级及权重信息等信息，不需要新生成针对该特性的轨迹线，也无需对状态机进行调整，系统可拓展性高。
本申请的实施例提供了一种智能驾驶装置，包括：处理器以及用于存储处理器可执行指令的存储器；其中，所述处理器被配置为执行所述指令时实现上述智能驾驶方法。
图17示出根据本申请一实施例的另一种智能驾驶装置的结构示意图，如图17所示，该智能驾驶装置可以包括：至少一个处理器1801，通信线路1802，存储器1803以及至少一个通信接口1804。
处理器1801可以是一个通用中央处理器(central processing unit，CPU)，微处理器，特定应用集成电路(application-specific integrated circuit，ASIC)，或一个或多个用于控制本申请方案程序执行的集成电路。
通信线路1802可包括一通路，在上述组件之间传送信息。
通信接口1804，使用任何收发器一类的装置，用于与其他设备或通信网络通信，如以太网，RAN，无线局域网(wireless local area networks，WLAN)等。
存储器1803可以是只读存储器(read-only memory，ROM)或可存储静态信息和指令的其他类型的静态存储设备，随机存取存储器(random access memory，RAM)或者可存储信息和指令的其他类型的动态存储设备，也可以是电可擦可编程只读存储器(electrically erasable programmable read-only memory，EEPROM)、只读光盘(compact disc read-only memory，CD-ROM)或其他光盘存储、光碟存储(包括压缩光碟、激光碟、光碟、数字通用光碟、蓝光光碟等)、磁盘存储介质或者其他磁存储设备、或者能够用于携带或存储具有指令或数据结构形式的期望的程序代码并能够由计算机存取的任何其他介质，但不限于此。存储器可以是独立存在，通过通信线路1802与处理器相连接。存储器也可以和处理器集成在一起。本申请实施例提供的存储器通常可以具有非易失性。其中，存储器1803用于存储执行本申请方案的计算机执行指令，并由处理器1801来控制执行。处理器1801用于执行存储器1803中存储的计算机执行指令，从而实现本申请上述实施例中提供的方法。
可选的，本申请实施例中的计算机执行指令也可以称之为应用程序代码。
在具体实现中，作为一种实施例，处理器1801可以包括一个或多个CPU，例如图17中的CPU0和CPU1。
在具体实现中，作为一种实施例，智能驾驶装置可以包括多个处理器，例如图17中的处理器1801和处理器1807。这些处理器中的每一个可以是一个单核(single-CPU)处理器，也可以是一个多核(multi-CPU)处理器。这里的处理器可以指一个或多个设备、电路、和/或用于处理数据(例如计算机程序指令)的处理核。
在具体实现中，作为一种实施例，智能驾驶装置还可以包括输出设备1805和输入设备1806。输出设备1805和处理器1801通信，可以以多种方式来显示信息。例如，输出设备1805可以是液晶显示器(liquid crystal display，LCD)，发光二级管(light emitting diode，LED)显示设备，阴极射线管(cathode ray tube，CRT)显示设备，或投影仪(projector)等。输入设备1806和处理器1801通信，可以以多种方式接收用户的输入。例如，输入设备1806可以是鼠标、键盘、触摸屏设备或传感设备等。
本申请实施例还提供一种高级驾驶辅助系统，应用于无人驾驶或智能驾驶中，其包含至少一个本申请上述实施例提到的智能驾驶装置，还可以包括摄像头或雷达等其他传感器中的至少一个，该传感器用于感知环境信息，该系统内的至少一个装置可以集成为一个整机或设备，或者该系统内的至少一个装置也可以独立设置为元件或装置。
本申请实施例还提供一种车辆，所述车辆包括至少一个本申请上述实施例提到的智能驾驶装置或上述任一系统。
本申请的实施例提供了一种非易失性计算机可读存储介质，其上存储有计算机程序指令，所述计算机程序指令被处理器执行时实现上述方法。
本申请的实施例提供了一种计算机程序产品，包括计算机可读代码，或者承载有计算机可读代码的非易失性计算机可读存储介质，当所述计算机可读代码在电子设备的处理器中运行时，所述电子设备中的处理器执行上述方法。
计算机可读存储介质可以是可以保持和存储由指令执行设备使用的指令的有形设备。计算机可读存储介质例如可以是――但不限于――电存储设备、磁存储设备、光存储设备、电磁存储设备、半导体存储设备或者上述的任意合适的组合。计算机可读存储介质的更具体的例子(非穷举的列表)包括：便携式计算机盘、硬盘、随机存取存储器(Random Access Memory，RAM)、只读存储器(Read Only Memory，ROM)、可擦式可编程只读存储器(Electrically Programmable Read-Only-Memory，EPROM或闪存)、静态随机存取存储器(Static Random-Access Memory，SRAM)、便携式压缩盘只读存储器(Compact Disc Read-Only Memory，CD-ROM)、数字多功能盘(Digital Video Disc，DVD)、记忆棒、软盘、机械编码设备、例如其上存储有指令的打孔卡或凹槽内凸起结构、以及上述的任意合适的组合。
这里所描述的计算机可读程序指令或代码可以从计算机可读存储介质下载到各个计算/处理设备，或者通过网络、例如因特网、局域网、广域网和/或无线网下载到外部计算机或外部存储设备。网络可以包括铜传输电缆、光纤传输、无线传输、路由器、防火墙、交换机、网关计算机和/或边缘服务器。每个计算/处理设备中的网络适配卡或者网络接口从网络接收计算机可读程序指令，并转发该计算机可读程序指令，以供存储在各个计算/处理设备中的计算机可读存储介质中。
用于执行本申请操作的计算机程序指令可以是汇编指令、指令集架构(Instruction Set Architecture，ISA)指令、机器指令、机器相关指令、微代码、固件指令、状态设 置数据、或者以一种或多种编程语言的任意组合编写的源代码或目标代码，所述编程语言包括面向对象的编程语言—诸如Smalltalk、C++等，以及常规的过程式编程语言—诸如“C”语言或类似的编程语言。计算机可读程序指令可以完全地在用户计算机上执行、部分地在用户计算机上执行、作为一个独立的软件包执行、部分在用户计算机上部分在远程计算机上执行、或者完全在远程计算机或服务器上执行。在涉及远程计算机的情形中，远程计算机可以通过任意种类的网络—包括局域网(Local Area Network，LAN)或广域网(Wide Area Network，WAN)—连接到用户计算机，或者，可以连接到外部计算机(例如利用因特网服务提供商来通过因特网连接)。在一些实施例中，通过利用计算机可读程序指令的状态信息来个性化定制电子电路，例如可编程逻辑电路、现场可编程门阵列(Field-Programmable Gate Array，FPGA)或可编程逻辑阵列(Programmable Logic Array，PLA)，该电子电路可以执行计算机可读程序指令，从而实现本申请的各个方面。
这里参照根据本申请实施例的方法、装置(系统)和计算机程序产品的流程图和/或框图描述了本申请的各个方面。应当理解，流程图和/或框图的每个方框以及流程图和/或框图中各方框的组合，都可以由计算机可读程序指令实现。
这些计算机可读程序指令可以提供给通用计算机、专用计算机或其它可编程数据处理装置的处理器，从而生产出一种机器，使得这些指令在通过计算机或其它可编程数据处理装置的处理器执行时，产生了实现流程图和/或框图中的一个或多个方框中规定的功能/动作的装置。也可以把这些计算机可读程序指令存储在计算机可读存储介质中，这些指令使得计算机、可编程数据处理装置和/或其他设备以特定方式工作，从而，存储有指令的计算机可读介质则包括一个制造品，其包括实现流程图和/或框图中的一个或多个方框中规定的功能/动作的各个方面的指令。
也可以把计算机可读程序指令加载到计算机、其它可编程数据处理装置、或其它设备上，使得在计算机、其它可编程数据处理装置或其它设备上执行一系列操作步骤，以产生计算机实现的过程，从而使得在计算机、其它可编程数据处理装置、或其它设备上执行的指令实现流程图和/或框图中的一个或多个方框中规定的功能/动作。
附图中的流程图和框图显示了根据本申请的多个实施例的装置、系统、方法和计算机程序产品的可能实现的体系架构、功能和操作。在这点上，流程图或框图中的每个方框可以代表一个模块、程序段或指令的一部分，所述模块、程序段或指令的一部分包含一个或多个用于实现规定的逻辑功能的可执行指令。在有些作为替换的实现中，方框中所标注的功能也可以以不同于附图中所标注的顺序发生。例如，两个连续的方框实际上可以基本并行地执行，它们有时也可以按相反的顺序执行，这依所涉及的功能而定。
也要注意的是，框图和/或流程图中的每个方框、以及框图和/或流程图中的方框的组合，可以用执行相应的功能或动作的硬件(例如电路或ASIC(Application Specific Integrated Circuit，专用集成电路))来实现，或者可以用硬件和软件的组合，如固件等来实现。
尽管在此结合各实施例对本发明进行了描述，然而，在实施所要求保护的本发明过程中，本领域技术人员通过查看所述附图、公开内容、以及所附权利要求书，可理解并实现所述公开实施例的其它变化。在权利要求中，“包括”(comprising)一词不排除其他组成部分或步骤，“一”或“一个”不排除多个的情况。单个处理器或其它单元可以实现权利 要求中列举的若干项功能。相互不同的从属权利要求中记载了某些措施，但这并不表示这些措施不能组合起来产生良好的效果。
以上所述，仅为本发明的具体实施方式，但本发明的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本发明揭露的技术范围内，可轻易想到变化或替换，都应涵盖在本发明的保护范围之内。因此，本发明的保护范围应以所述权利要求的保护范围为准。

Claims (22)

  1. 一种智能驾驶方法，其特征在于，所述方法包括：
    获取当前时刻自车周围的环境信息及当前时刻自车的状态信息；
    根据所述环境信息及所述状态信息，确定当前时刻的控制点；
    根据所述控制点、历史时刻的控制轨迹及所述状态信息，生成当前时刻的控制轨迹，所述控制轨迹表示用于导引自车行驶的轨迹。
  2. 根据权利要求1所述的方法，其特征在于，所述根据所述环境信息及所述状态信息，确定当前时刻的控制点，包括：
    根据所述环境信息中各第一对象的预设优先级信息及所述状态信息，依次确定所述当前时刻的控制点。
  3. 根据权利要求2所述的方法，其特征在于，所述第一对象包括：自车道的车道线、目标车辆、邻车道中车辆；其中，所述自车道的车道线、所述目标车辆、所述邻车道中车辆的优先级依次降低；
    所述根据所述环境信息中各第一对象的预设优先级信息及所述状态信息，依次确定所述当前时刻的控制点，包括：
    在所述自车道的车道线存在的情况下，将所述自车道的车道线上预设距离对应的、且符合预设条件的位置点作为当前时刻的控制点；所述预设距离基于所述状态信息确定；
    如果所述自车道的车道线不存在，则在所述目标车辆存在的情况下，将所述目标车辆对应的、且符合预设条件的位置点作为当前时刻的控制点；
    如果所述自车道的车道线不存在，且所述目标车辆不存在，则在所述邻车道中车辆存在的情况下，将所述邻车道中车辆在自车道投影的、且符合预设条件的位置点作为当前时刻的控制点。
  4. 根据权利要求2或3所述的方法，其特征在于，所述根据所述环境信息中各第一对象的预设优先级信息及所述状态信息，依次确定所述当前时刻的控制点，包括：
    获取优先级最高的第一对象对应的位置点；
    判断预设区域内是否存在第二对象，所述第二对象为环境信息中自车的避让对象；
    在所述第二对象存在时，对所述第一对象对应的位置点进行偏移处理，并将偏移后的位置点作为当前时刻的控制点。
  5. 根据权利要求4所述的方法，其特征在于，所述第二对象包括自车道的硬边界；所述预设区域为以第一对象对应的位置点为中心预设大小的区域；
    所述在所述第二对象存在时，对所述第一对象对应的位置点进行偏移处理，并将偏移后的位置点作为当前时刻的控制点，包括：
    在所述自车道的硬边界存在时，对所述第一对象对应的位置点进行偏移处理，以使所述硬边界在以所述偏移后的位置点为中心预设大小的区域之外，且所述偏移后的位置点在自车道内；
    将偏移后的位置点作为当前时刻的控制点。
  6. 根据权利要求4所述的方法，其特征在于，所述第二对象包括障碍物；所述预设区域为所述自车道中自车前方预设大小的区域；
    所述在所述第二对象存在时，对所述第一对象对应的位置点进行偏移处理，并将偏移后的位置点作为当前时刻的控制点，包括：
    在所述障碍物存在时，根据所述障碍物对应的权重信息，对所述第一对象对应的位置点进行偏移处理，以使所述障碍物在以所述偏移后的位置点为中心预设大小的区域之外，且所述偏移后的位置点在自车道内；其中，所述权重信息表示自车避让该障碍物的程度；
    将偏移后的位置点作为当前时刻的控制点。
  7. 根据权利要求6所述的方法，其特征在于，所述权重信息的数值至少由车头时距THW、碰撞时间TTC、侵入宽度IVW中的一个确定，其中，所述权重信息的数值与所述THW成负相关，与所述TTC成负相关，与所述IVW成正相关。
  8. 根据权利要求1所述的方法，其特征在于，所述根据所述控制点、历史时刻的控制轨迹及所述状态信息，生成当前时刻的控制轨迹，包括：
    根据所述控制点、历史时刻的控制轨迹及所述状态信息，利用卡尔曼滤波器，生成所述当前时刻的控制轨迹。
  9. 一种智能驾驶装置，其特征在于，所述装置包括：
    获取模块，用于获取当前时刻自车周围的环境信息及当前时刻自车的状态信息；
    确定模块，用于根据所述环境信息及所述状态信息，确定当前时刻的控制点；
    生成模块，用于根据所述控制点、历史时刻的控制轨迹及所述状态信息，生成当前时刻的控制轨迹，所述控制轨迹表示用于导引自车行驶的轨迹。
  10. 根据权利要求9所述的装置，其特征在于，所述确定模块，还用于：根据所述环境信息中各第一对象的预设优先级信息及所述状态信息，依次确定所述当前时刻的控制点。
  11. 根据权利要求10所述的装置，其特征在于，所述第一对象包括：自车道的车道线、目标车辆、邻车道中车辆；其中，所述自车道的车道线、所述目标车辆、所述邻车道中车辆的优先级依次降低；
    所述确定模块，还用于：在所述自车道的车道线存在的情况下，将所述自车道的车道线上预设距离对应的、且符合预设条件的位置点作为当前时刻的控制点；所述预设距离基于所述状态信息确定；如果所述自车道的车道线不存在，则在所述目标车辆存在的情况下，将所述目标车辆对应的、且符合预设条件的位置点作为当前时刻的控制点；如果所述自车道的车道线不存在，且所述目标车辆不存在，则在所述邻车道中车辆存在的情况下，将所述邻车道中车辆在自车道投影的、且符合预设条件的位置点作为当前时刻的控制点。
  12. 根据权利要求10或11所述的装置，其特征在于，所述确定模块，还用于：获取优先级最高的第一对象对应的位置点；判断预设区域内是否存在第二对象，所述第二对象为环境信息中自车的避让对象；在所述第二对象存在时，对所述第一对象对应的位置点进行偏移处理，并将偏移后的位置点作为当前时刻的控制点。
  13. 根据权利要求12所述的装置，其特征在于，所述第二对象包括自车道的硬边界；所述预设区域为以第一对象对应的位置点为中心预设大小的区域；
    所述确定模块，还用于：在所述自车道的硬边界存在时，对所述第一对象对应的位置点进行偏移处理，以使所述硬边界在以所述偏移后的位置点为中心预设大小的区域之外，且所述偏移后的位置点在自车道内；将偏移后的位置点作为当前时刻的控制点。
  14. 根据权利要求12所述的装置，其特征在于，所述第二对象包括障碍物；所述预设区域为所述自车道中自车前方预设大小的区域；
    所述确定模块，还用于：在所述障碍物存在时，根据所述障碍物对应的权重信息，对所述第一对象对应的位置点进行偏移处理，以使所述障碍物在以所述偏移后的位置点为中心预设大小的区域之外，且所述偏移后的位置点在自车道内；其中，所述权重信息表示自车避让该障碍物的程度；将偏移后的位置点作为当前时刻的控制点。
  15. 根据权利要求14所述的装置，其特征在于，所述权重信息的数值至少由车头时距THW、碰撞时间TTC、侵入宽度IVW中的一个确定，其中，所述权重信息的数值与所述THW成负相关，与所述TTC成负相关，与所述IVW成正相关。
  16. 根据权利要求9所述的装置，其特征在于，所述生成模块，还用于：根据所述控制点、历史时刻的控制轨迹及所述状态信息，利用卡尔曼滤波器，生成所述当前时刻的控制轨迹。
  17. 一种高级驾驶辅助系统ADAS，其特征在于，包括：预处理层、规划层、决策层、控制层；
    其中，所述预处理层，用于生成当前时刻自车周围的环境信息；
    所述规划层，用于根据所述环境信息及当前时刻自车的状态信息，确定当前时刻的控制点；根据所述控制点、历史时刻的控制轨迹及所述状态信息，生成当前时刻的控制轨迹；所述控制轨迹表示用于导引自车行驶的轨迹；
    所述决策层，用于对所述控制轨迹是否适用于当前时刻自车的工作状态进行判断；
    所述控制层，用于在所述控制轨迹适用于当前时刻自车的工作状态的情况下，基于所述控制轨迹，生成当前时刻的控制信号，所述控制信号用于车辆的辅助驾驶。
  18. 根据权利要求17所示的系统，其特征在于，所述规划层还用于：根据所述环境信息中各第一对象的预设优先级信息及所述状态信息，依次确定所述当前时刻的控制点。
  19. 根据权利要求18所示的系统，其特征在于，所述规划层还用于：获取优先级最高的第一对象对应的位置点；判断预设区域内是否存在第二对象，所述第二对象为环境信息中自车的避让对象；在所述第二对象存在时，对所述第一对象对应的位置点进行偏移处理，并将偏移后的位置点作为当前时刻的控制点。
  20. 一种智能驾驶装置，其特征在于，包括：
    处理器；
    用于存储处理器可执行指令的存储器；
    其中，所述处理器被配置为执行所述指令时实现权利要求1-8中任意一项所述的方法。
  21. 一种非易失性计算机可读存储介质，其上存储有计算机程序指令，其特征在于，所述计算机程序指令被处理器执行时实现权利要求1-8中任意一项所述的方法。
  22. 一种包含指令的计算机程序产品，其特征在于，当其在计算机上运行时，使得所述计算机执行如权利要求1-8中任意一项所述的方法。
PCT/CN2021/078497 2021-03-01 2021-03-01 一种智能驾驶方法、装置、存储介质及计算机程序 WO2022183329A1 (zh)

Priority Applications (3)

Application Number Priority Date Filing Date Title
PCT/CN2021/078497 WO2022183329A1 (zh) 2021-03-01 2021-03-01 一种智能驾驶方法、装置、存储介质及计算机程序
EP21928420.5A EP4296133A4 (en) 2021-03-01 2021-03-01 INTELLIGENT DRIVE METHOD AND DEVICE AS WELL AS STORAGE MEDIUM AND COMPUTER PROGRAM
CN202180000458.5A CN113165670A (zh) 2021-03-01 2021-03-01 一种智能驾驶方法、装置、存储介质及计算机程序

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2021/078497 WO2022183329A1 (zh) 2021-03-01 2021-03-01 一种智能驾驶方法、装置、存储介质及计算机程序

Publications (1)

Publication Number Publication Date
WO2022183329A1 true WO2022183329A1 (zh) 2022-09-09

Family

ID=76875968

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/078497 WO2022183329A1 (zh) 2021-03-01 2021-03-01 一种智能驾驶方法、装置、存储介质及计算机程序

Country Status (3)

Country Link
EP (1) EP4296133A4 (zh)
CN (1) CN113165670A (zh)
WO (1) WO2022183329A1 (zh)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114090560A (zh) * 2021-11-19 2022-02-25 湖北亿咖通科技有限公司 一种车道的中心线生成方法、装置、设备及存储介质
CN115966066A (zh) * 2023-03-15 2023-04-14 中国安全生产科学研究院 一种矿山车辆安全运行监测预警方法、装置和系统
CN117272690A (zh) * 2023-11-21 2023-12-22 中汽智联技术有限公司 自动驾驶车辆危险切入场景提取方法、设备及介质
CN117261903A (zh) * 2023-11-21 2023-12-22 杭州鉴智机器人科技有限公司 一种自动驾驶车辆的变道方法及装置

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114228741A (zh) * 2021-11-16 2022-03-25 智道网联科技(北京)有限公司 用于自动驾驶车辆的车道保持方法及装置
WO2023092451A1 (zh) * 2021-11-26 2023-06-01 华为技术有限公司 预测可行驶车道的方法和装置
CN114506346B (zh) * 2022-03-24 2023-11-21 重庆长安汽车股份有限公司 一种用于自动驾驶的速度控制方法及存储介质
CN116125897A (zh) * 2023-03-02 2023-05-16 重庆长安汽车股份有限公司 自动驾驶决策方法、装置、电子设备及存储介质

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107433946A (zh) * 2016-05-27 2017-12-05 现代自动车株式会社 考虑优先级的用于控制变道的装置和方法
CN109131326A (zh) * 2018-08-08 2019-01-04 大连理工大学 一种具有换道辅助功能的自适应巡航控制器及其工作方法
CN110588653A (zh) * 2018-05-24 2019-12-20 通用汽车环球科技运作有限责任公司 自主车辆的控制系统、控制方法以及控制器
CN111133485A (zh) * 2017-08-23 2020-05-08 Uatc有限责任公司 自主交通工具的对象预测优先级排序系统和方法
EP3683782A1 (en) * 2019-01-18 2020-07-22 Honda Research Institute Europe GmbH Method for assisting a driver, driver assistance system, and vehicle including such driver assistance system
CN111538328A (zh) * 2020-04-03 2020-08-14 浙江工业大学 一种用于自主驾驶车辆避障轨迹规划与跟踪控制的优先级分层预测控制方法
CN112216146A (zh) * 2020-09-16 2021-01-12 浙江吉利控股集团有限公司 一种车辆风险预警方法、装置、设备及车辆

Family Cites Families (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2006070865A1 (ja) * 2004-12-28 2006-07-06 Kabushiki Kaisha Toyota Chuo Kenkyusho 車両運動制御装置
KR102481487B1 (ko) * 2018-02-27 2022-12-27 삼성전자주식회사 자율 주행 장치 및 그의 제어 방법
WO2019231456A1 (en) * 2018-05-31 2019-12-05 Nissan North America, Inc. Probabilistic object tracking and prediction framework
US11827241B2 (en) * 2018-10-29 2023-11-28 Motional Ad Llc Adjusting lateral clearance for a vehicle using a multi-dimensional envelope
CN111383474B (zh) * 2018-12-29 2022-01-25 毫末智行科技有限公司 自动驾驶车辆的决策系统及方法
CN110239550B (zh) * 2019-05-28 2021-01-19 浙江吉利控股集团有限公司 一种车辆自动变道方法
CN110262486B (zh) * 2019-06-11 2020-09-04 北京三快在线科技有限公司 一种无人驾驶设备运动控制方法及装置
CN111907521B (zh) * 2020-06-15 2022-11-22 浙江吉利汽车研究院有限公司 一种自动驾驶车辆的横向控制方法、装置及存储介质
CN111856470A (zh) * 2020-07-20 2020-10-30 中国第一汽车股份有限公司 一种汽车雷达的目标筛选方法、装置、设备和存储介质

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107433946A (zh) * 2016-05-27 2017-12-05 现代自动车株式会社 考虑优先级的用于控制变道的装置和方法
CN111133485A (zh) * 2017-08-23 2020-05-08 Uatc有限责任公司 自主交通工具的对象预测优先级排序系统和方法
CN110588653A (zh) * 2018-05-24 2019-12-20 通用汽车环球科技运作有限责任公司 自主车辆的控制系统、控制方法以及控制器
CN109131326A (zh) * 2018-08-08 2019-01-04 大连理工大学 一种具有换道辅助功能的自适应巡航控制器及其工作方法
EP3683782A1 (en) * 2019-01-18 2020-07-22 Honda Research Institute Europe GmbH Method for assisting a driver, driver assistance system, and vehicle including such driver assistance system
CN111538328A (zh) * 2020-04-03 2020-08-14 浙江工业大学 一种用于自主驾驶车辆避障轨迹规划与跟踪控制的优先级分层预测控制方法
CN112216146A (zh) * 2020-09-16 2021-01-12 浙江吉利控股集团有限公司 一种车辆风险预警方法、装置、设备及车辆

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP4296133A4 *

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114090560A (zh) * 2021-11-19 2022-02-25 湖北亿咖通科技有限公司 一种车道的中心线生成方法、装置、设备及存储介质
CN114090560B (zh) * 2021-11-19 2023-12-26 亿咖通(湖北)技术有限公司 一种车道的中心线生成方法、装置、设备及存储介质
CN115966066A (zh) * 2023-03-15 2023-04-14 中国安全生产科学研究院 一种矿山车辆安全运行监测预警方法、装置和系统
CN117272690A (zh) * 2023-11-21 2023-12-22 中汽智联技术有限公司 自动驾驶车辆危险切入场景提取方法、设备及介质
CN117261903A (zh) * 2023-11-21 2023-12-22 杭州鉴智机器人科技有限公司 一种自动驾驶车辆的变道方法及装置
CN117272690B (zh) * 2023-11-21 2024-02-23 中汽智联技术有限公司 自动驾驶车辆危险切入场景提取方法、设备及介质

Also Published As

Publication number Publication date
CN113165670A (zh) 2021-07-23
EP4296133A4 (en) 2024-04-24
EP4296133A1 (en) 2023-12-27

Similar Documents

Publication Publication Date Title
WO2022183329A1 (zh) 一种智能驾驶方法、装置、存储介质及计算机程序
US11462022B2 (en) Traffic signal analysis system
US11003183B2 (en) Driving scene based path planning for autonomous driving vehicles
EP3342666B1 (en) Method and system for operating autonomous driving vehicles using graph-based lane change guide
US11117597B2 (en) Pedestrian interaction system for low speed scenes for autonomous vehicles
US11260855B2 (en) Methods and systems to predict object movement for autonomous driving vehicles
US11945434B2 (en) Delay decision making for autonomous driving vehicles in response to obstacles based on confidence level and distance
US10668925B2 (en) Driver intention-based lane assistant system for autonomous driving vehicles
US10489686B2 (en) Object detection for an autonomous vehicle
US10606277B2 (en) Speed optimization based on constrained smoothing spline for autonomous driving vehicles
EP3324332B1 (en) Method and system to predict vehicle traffic behavior for autonomous vehicles to make driving decisions
US10591926B2 (en) Smooth road reference for autonomous driving vehicles based on 2D constrained smoothing spline
CN111619560B (zh) 车辆控制方法及装置
US10807599B2 (en) Driving scenario based lane guidelines for path planning of autonomous driving vehicles
US10849543B2 (en) Focus-based tagging of sensor data
US10860868B2 (en) Lane post-processing in an autonomous driving vehicle
JP6574224B2 (ja) 車両制御装置、車両、車両制御方法およびプログラム
US10909377B2 (en) Tracking objects with multiple cues
CN111857118A (zh) 对停车轨迹分段以控制自动驾驶车辆停车
JP6658968B2 (ja) 運転支援方法及び運転支援装置
US11787440B2 (en) Lane boundary and vehicle speed based nudge decision
US11724717B2 (en) Implementation of dynamic cost function of self-driving vehicles
JP2019045985A (ja) 車両制御装置、車両、車両制御方法およびプログラム
US20210303874A1 (en) A point cloud-based low-height obstacle detection system
US20210262819A1 (en) A mixed regular and open-space trajectory planning method for autonomous driving vehicle

Legal Events

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

Ref document number: 21928420

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 2021928420

Country of ref document: EP

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2021928420

Country of ref document: EP

Effective date: 20230920