WO2022016351A1

WO2022016351A1 - Method and apparatus for selecting driving decision

Info

Publication number: WO2022016351A1
Application number: PCT/CN2020/103181
Authority: WO
Inventors: 李帅君
Original assignee: 华为技术有限公司
Priority date: 2020-07-21
Filing date: 2020-07-21
Publication date: 2022-01-27
Also published as: CN112512887A; CN112512887B

Abstract

A method and apparatus for selecting a driving decision, which relate to the fields of smart cars, connected vehicles and automatic driving, and are used for enlarging the perception range available when selecting a driving decision and planning a traveling path for a vehicle, thereby improving the driving safety and stability of the vehicle, and improving the user experience. The method comprises: acquiring perception information collected by a sensor configured on a vehicle (203); selecting, from at least one range interval, a first range interval that matches the perception information (204), wherein the at least one range interval is obtained by dividing a monitoring range of the sensor on the basis of output information from the sensor, and each range interval corresponds to at least one driving decision; in conjunction with the speed of the vehicle, selecting a driving decision for the vehicle from at least one driving decision corresponding to the first range interval (205); and controlling the vehicle to travel according to the driving decision for the vehicle (206).

Description

A driving decision selection method and device

technical field

The present application relates to the field of automobiles, and in particular, to a driving decision selection method and device.

Background technique

Decision-making and planning are an important component of autonomous driving technology, such as global path planning, behavioral decision-making, or local path planning. At this stage, the upstream perception module usually outputs the location information of the environment and some key objects, and the decision and planning module generates the control target for the vehicle, such as the reference path or reference trajectory, and outputs it to the downstream control link. Complete the closed-loop control of the vehicle.

However, when planning a driving path, it is usually based on deterministic perception information, that is, perception information with a confidence level higher than a certain value, which will result in a limited range of perception.

SUMMARY OF THE INVENTION

Embodiments of the present application provide a driving decision selection method and device, which are used to expand the available perception range when selecting a driving decision for a vehicle and planning a driving path, improve the driving safety and stability of the vehicle, and improve user experience.

In view of this, in the first aspect, the present application provides a driving decision selection method, including: first, acquiring perception information collected by sensors configured on the vehicle; Range interval, the at least one range interval is obtained by dividing the detection range of the sensor based on the output information of the sensor, and the output information of the sensor includes the distance result of the detected target, the confidence level corresponding to the distance result, or the distance result and the distance result. At least one of the relationships between the confidence levels corresponding to the distance results, and each range interval has at least one corresponding driving decision; then, combined with the speed of the vehicle, select the vehicle's driving decision from the at least one driving decision corresponding to the first range interval. Driving decisions, and control the vehicle to drive according to the driving decisions of the vehicle.

In the embodiment of the present application, before the driving decision is selected, the detection range of the sensor has been divided into at least one range section, and each range section has a corresponding distance range. In the process of selecting a driving decision, at least one driving decision corresponding to each range interval can be used to select the driving decision of the vehicle, such as acceleration, deceleration, maintaining the speed or changing lanes, and control the vehicle to drive according to the driving decision. Generally, the farther away the object is detected by the sensor, the lower the confidence level. Therefore, instead of using only perceptual information with high confidence to plan a driving path, in the driving decision selection method provided by the present application, a range interval is divided for the detection range of the sensor, and one or more range intervals obtained cover the detection range of the sensor. Scope. In the process of choosing a driving decision, even if the perceived information includes a long distance or a low degree of confidence, the farther perceived distance can be used to select a driving decision, so that the vehicle can evade further obstacles in advance , which currently increases the available perception range when deciding the driving decision of the vehicle, improves the driving safety of the vehicle, and improves the user experience. In addition, to avoid obstacles that are farther away from the vehicle in advance, decisions such as acceleration or deceleration can be made in advance to avoid the violent acceleration or deceleration caused by the vehicle being too close, so that the driving process of the vehicle is smoother and the user experience is improved. .

In a possible implementation, each range interval corresponds to a distance range, the perception information includes a first distance between the obstacle detected by the sensor and the vehicle, and a first distance matching the perception information is selected from at least one range interval The range interval, which may include: matching the first distance with the distance range corresponding to the aforementioned at least one range interval, so as to know that the first distance is within the distance range corresponding to the first range interval, and then filter out the matching with the first distance. the first range interval.

In the embodiment of the present application, each range section has a corresponding driving decision, so that a range section that matches the range section can be selected according to the distance between the vehicle and the object, and the driving of the vehicle can be selected according to one or more driving decisions corresponding to the range section Therefore, the driving decision of the vehicle can be selected based on a larger perception range, and then the vehicle driving path can be planned based on the larger perception range.

In a possible implementation manner, each range interval corresponds to a confidence range, and the confidence range corresponding to the at least one range interval covers the confidence of the information detected by the sensor within the detection range. The information also includes a first confidence level, and the first confidence level is used to indicate the accuracy of the first distance; the above-mentioned selection of the first range interval that matches the perceptual information from the at least one range interval may include: The degree of confidence is matched with the confidence degree range corresponding to the aforementioned at least one range interval, so as to know that the first confidence degree is within the confidence degree range corresponding to the first range interval, and then the first range interval is filtered out.

In the embodiments of the present application, a matching range interval may be selected based on the confidence level included in the perception information, and then a driving decision of the vehicle may be selected from at least one driving decision in the range interval. The vehicle's driving decisions can be selected even in low-confidence scenarios. Then, the driving decision is selected for a longer distance or a distance with a lower confidence, so that the vehicle can deal with the obstacles in the longer distance in advance, such as decelerating in advance or changing lanes in advance, etc., to improve the safety of the vehicle.

In a possible implementation manner, the first confidence level included in the perception information is obtained according to the detection range of the sensor and the distance between the sensor and the obstacle. Among them, the sensor is usually arranged in the vehicle, and the distance between the sensor and the obstacle can be understood as the distance between the vehicle and the sensor. Generally, the farther the distance between the vehicle and the sensor is, the lower the first confidence level is, and the closer the distance between the vehicle and the sensor is, the higher the first confidence level is.

In a possible implementation manner, before acquiring the perception information, the driving decision selection method provided by the present application further includes: dividing the detection range of the sensor to obtain at least one distance range, the at least one distance range being equal to the at least one range interval A correspondence.

Therefore, in the embodiment of the present application, the detection range of the sensor can be divided to obtain the distance range corresponding to each range interval, so that the distance range corresponding to the aforementioned at least one range interval can cover the detection range of the sensor, which is equivalent to increasing the The range of perception available when choosing driving decisions. In a possible implementation manner, before acquiring the sensing information, dividing the detection range of the sensor to obtain at least one distance range may include: dividing the range of confidence levels detectable by the sensor within the detection range to obtain at least one The confidence range, the confidence range corresponding to the at least one range interval, covers the confidence of the information detected by the sensor within the detection range. Then, according to the relationship between the distance result and the confidence level corresponding to the distance result, the distance range corresponding to each confidence level range is calculated to obtain at least one distance range.

In the embodiment of the present application, the detectable confidence range of the sensor is divided into at least one confidence range, each confidence range has a corresponding distance range, and each range interval is set with at least one driving decision. Therefore, when choosing a driving decision, the driving decision can be selected according to the distance and confidence included in the perception information, and then the driving path of the vehicle can be planned. Planning the driving path is equivalent to expanding the available perception range when planning the driving path, improving the driving safety and stability of the vehicle, and improving the user experience. It can be understood that, compared with only using perception information with a confidence higher than a certain value to plan the driving path, the range interval is delimited in advance in this application, so that when planning the driving path later, a larger perception range can be used to plan driving. Therefore, an appropriate driving decision can be selected in advance for the obstacles in front of the vehicle, so as to quickly and safely plan a driving path with higher safety and stability, and improve the user experience.

In a possible implementation manner, according to the relationship between the distance result and the confidence level corresponding to the distance result, at least one confidence level range corresponding to at least one distance range is determined, and the at least one confidence level range is used to start from the at least one range. In the interval, filter the range interval that matches the perception information.

In a possible implementation manner, before acquiring the perception information, the driving decision selection method provided by the present application may further include: dividing the detection range of the sensor to obtain at least one distance range, that is, the at least one distance range covers the sensor Then, according to at least one distance range and the relationship between the distance result and the confidence level corresponding to the distance result, at least one confidence level corresponding to at least one distance range is obtained, and each range interval corresponds to a confidence level range and distance range.

In the embodiment of the present application, the confidence range can be divided first, and then the distance range corresponding to each confidence degree is calculated according to the relationship between the distance result and the confidence degree corresponding to the distance result, that is, a range interval has a corresponding distance interval and confidence degree Therefore, the driving decision can be selected according to the distance or confidence in the future, which increases the available perception range when choosing the driving decision, improves the safety and stability of the vehicle, and improves the user experience.

In a possible implementation, if at least one range interval is obtained by dividing the monitoring range according to the distance result output by the sensor, the first distance included in the sensing information and the distance range corresponding to each range interval can be directly performed. Match, filter out the range interval that matches the first distance.

In another possible implementation, if at least one range interval is obtained according to the confidence level corresponding to the distance result output by the sensor, and each range interval corresponds to a confidence level range, the first range included in the sensing information can be divided into A confidence level is matched with a distance range corresponding to each range interval, so as to filter out a range interval matching the first confidence level.

In another possible implementation, if at least one range interval is divided according to the distance result, and the perception information includes the relationship between the first confidence level and the aforementioned distance result and the confidence level corresponding to the distance result, then The first distance corresponding to the first confidence degree can be calculated according to the relationship between the distance result and the confidence degree corresponding to the distance result, and then the first distance is matched with the distance range corresponding to each range interval, so as to obtain the The range interval in which the first distance matches.

In a possible implementation manner, the driving decision selection method provided by the present application may further include: acquiring historical distance information collected by a sensor, historical distance information, and corresponding confidence; obtaining a distance result according to the historical distance information and corresponding confidence The relationship between the confidence levels corresponding to the distance results.

In the embodiment of the present application, the relationship between the distance result included in the output information of the sensor and the confidence degree corresponding to the distance result can be counted according to the historical distance information and the corresponding confidence degree collected by the sensor, so as to carry out the follow-up The division of the range interval, so as to determine the confidence range and distance range corresponding to each range interval.

In a possible implementation, when selecting the driving decision of the vehicle, the relative speed of the vehicle and the obstacle can be calculated according to the speed of the vehicle and the speed of the obstacle, and the speed of the obstacle can be obtained according to a period of time. Then, the driving decision of the vehicle is selected from at least one driving decision corresponding to the first range interval in combination with the relative speed.

In the embodiment of the present application, when selecting the driving decision, the driving decision of the vehicle can be selected in combination with the relative speed between the vehicle and the obstacle, so as to select the driving decision of the vehicle more accurately, and further improve the driving safety of the vehicle.

In a possible implementation, at least one driving decision corresponding to each range interval is determined according to an application scenario, and the application scenario may include, but is not limited to, scenarios such as automatic cruise, car following, or automatic parking. In a second aspect, the present application provides a driving decision selection device, which has the function of implementing the driving decision selection method of the first aspect. This function can be implemented by hardware or by executing corresponding software by hardware. The hardware or software includes one or more modules corresponding to the above functions.

In a second aspect, the present application provides a driving decision selection device, which has the function of implementing the driving decision selection method of the first aspect. This function can be implemented by hardware or by executing corresponding software by hardware. The hardware or software includes one or more modules corresponding to the above functions.

In a third aspect, the present application provides a driving decision selection method, comprising: acquiring at least one range interval, each range interval in the at least one range interval corresponds to a confidence range and a distance range, and the distance included in the at least one range interval The range covers the detection range of the sensor, and the at least one confidence range covers the confidence level of the information detected by the sensor within the detection range; decision, each range section and at least one driving decision corresponding to each range section are used to select the driving decision of the vehicle, and the driving decision of the vehicle is used to generate the driving path of the vehicle.

In the embodiment of the present application, the detection range of the sensor is divided into one or more distance ranges, each distance range has a corresponding confidence range, a range range corresponds to a distance range and a confidence range, and each range range corresponds to at least A driving decision. In the process of choosing a driving decision, you can use confidence or distance for matching, and set driving decisions for each range according to fixed rules, such as acceleration, deceleration, maintaining the speed or changing lanes, etc. Driving decision driving. Generally, the farther away the object is detected by the sensor, the lower the confidence level. Therefore, in the driving decision selection method provided by the present application, in the process of selecting a driving decision, the driving decision can be selected by using a perceived farther distance, which can be understood as a driving decision can be selected for objects with a longer distance, so that the vehicle Further obstacles can be avoided in advance, which improves the driving safety of the vehicle and improves the user experience. In addition, the obstacle that is farther away from the vehicle can be avoided in advance, and decisions such as acceleration or deceleration can be made in advance, so that the driving process of the vehicle is smoother and the user experience is improved.

In a possible implementation manner, acquiring at least one range interval may include: first, dividing the detection range of the sensor to obtain at least one distance range, and corresponding to the distance result according to the distance result included in the output information of the sensor The relationship between the confidence levels, calculate the confidence range corresponding to each distance range, and a range interval corresponds to a confidence range and a distance range.

Therefore, in the embodiments of the present application, the detection range of the sensor can be divided based on the distance between the vehicle and the object to obtain one or more distance ranges, and at least one range interval can be obtained according to the confidence range corresponding to each distance range. That is, a range interval has a corresponding distance interval and a confidence interval, so that the driving decision can be selected based on a longer distance or a lower confidence in the future, which is equivalent to increasing the available perception range when choosing a driving decision and improving the safety of the vehicle. and stability to improve user experience.

In a possible implementation manner, acquiring at least one range interval may include: dividing the range of confidence levels that can be detected by the sensor to obtain at least one confidence level range, and according to the distance result included in the output information of the sensor and the range The relationship between the confidence levels corresponding to the distance results is calculated. The distance range corresponding to each confidence level range is calculated. A range interval corresponds to a confidence range and a distance range.

Therefore, in the embodiments of the present application, one or more confidence ranges may be obtained by dividing based on the confidence. Usually, there is a correspondence between confidence and distance, each confidence range has a corresponding distance range, and the confidence range and distance range can constitute one or more range intervals obtained by dividing the sensing range of the sensor.

In a possible implementation manner, after setting at least one driving decision corresponding to each range interval in the at least one range interval, the above method further includes: acquiring sensing information collected by a sensor set in the vehicle, the sensing The information may include information on obstacles, such as the first distance between the vehicle and the obstacle; select a range interval that matches the perception information from at least one range interval, and combine the speed of the vehicle to select a range interval from at least one of the first range interval. In a driving decision, the driving decision of the vehicle is selected, and the vehicle is controlled to travel according to the driving decision, so that the vehicle travels according to the driving decision.

Therefore, in the embodiment of the present application, in the process of selecting the driving decision of the vehicle, the driving decision of the vehicle may be selected based on at least one driving decision corresponding to each range interval in the at least one range interval obtained by the foregoing division. Compared with only using perceptual information with a high degree of confidence to decide the driving decision of the vehicle, the distance range corresponding to at least one range interval provided in this application covers the detection range of the sensor, even if the perceptual information has a low degree of confidence, this Perceptual information is equivalent to selecting driving decisions based on a longer distance or lower confidence, so that driving decisions can be selected using a larger perception range, and then the driving path of the vehicle can be planned. In this way, a more accurate and safer driving path can be planned.

In a possible implementation, the perception information includes the distance between the obstacle and the vehicle, and according to the perception information and at least one driving decision corresponding to each range interval, selecting the driving decision of the vehicle may include: filtering out the obstacles and the vehicle The distance of is within the distance range in the first range interval; then, a driving decision of the vehicle is selected from at least one driving decision in the first range interval in combination with the speed of the vehicle.

Therefore, in the embodiment of the present application, each range section has a corresponding driving decision, so that a range section that matches the range section can be selected according to the distance of the object detected by the sensor, and one or more driving decisions corresponding to the range section can be selected. The driving decision of the vehicle is selected, so that the driving decision of the vehicle can be selected based on a larger perception range.

In a possible implementation, the perception information may include a confidence level; the above-mentioned driving decision for selecting the vehicle according to the perception information and at least one driving decision corresponding to each range interval may also include: if the perception information includes If the confidence level is within the confidence level range corresponding to the first range interval, a driving decision of the vehicle is selected from at least one driving decision in the first range interval in combination with the speed of the vehicle.

Therefore, in the embodiment of the present application, the corresponding range interval can be selected based on the confidence level included in the perception information, and then the driving decision of the vehicle can be selected. Or objects with lower confidence choose driving decisions to improve the safety of vehicle driving.

In a possible implementation, the relative speed of the vehicle relative to the obstacle can be calculated according to the speed of the vehicle, and then a driving decision of the vehicle is selected from at least one driving decision in the first range in combination with the relative speed.

Therefore, in the embodiment of the present application, the driving decision of the vehicle can be more accurately selected in combination with the speed of the vehicle and/or the relative speed of the vehicle and the obstacle, thereby further improving the safety of the driving of the vehicle.

In a possible implementation, the confidence level included in the perception information is related to the distance between the sensor and the obstacle. In general, the information perceived by the sensor increases with distance, and the confidence level decreases.

In a possible implementation manner, acquiring the relationship between the distance result included in the output information of the sensor and the confidence level corresponding to the distance result may include: acquiring historical distance information collected by the sensor, historical distance information and corresponding confidence degree; according to the relationship between the distance result included in the historical distance information and the output information of the corresponding confidence level sensor and the confidence level corresponding to the distance result.

Therefore, in the embodiment of the present application, the relationship between the distance result included in the output information of the sensor and the confidence degree corresponding to the distance result can be counted according to the historical distance information and the corresponding confidence degree collected by the sensor, to obtain Complete the subsequent division of the range interval.

In a fourth aspect, the present application provides a driving decision selection device, which has the function of implementing the driving decision selection method of the third aspect. This function can be implemented by hardware or by executing corresponding software by hardware. The hardware or software includes one or more modules corresponding to the above functions.

In a fifth aspect, an embodiment of the present application provides a driving decision selection device, the driving decision selection device has the function of implementing the driving decision selection method of the first aspect or the third aspect. This function can be implemented by hardware or by executing corresponding software by hardware. The hardware or software includes one or more modules corresponding to the above functions.

In a sixth aspect, an embodiment of the present application provides a driving decision and selection device, including: a processor and a memory, wherein the processor and the memory are interconnected through a line, and the processor invokes program codes in the memory to execute the first aspect or the first aspect. A function related to processing in the driving decision selection method shown in any one of the three aspects.

In a seventh aspect, an embodiment of the present application provides a driving decision and selection device, which may also be called a digital processing chip or a chip. The chip includes a processing unit and a communication interface. The processing unit obtains program instructions through the communication interface. The instructions are executed by a processing unit, and the processing unit is configured to perform processing-related functions in the first aspect, any optional implementation manner of the first aspect, the third aspect, or any optional implementation manner of the third aspect.

Optionally, the aforementioned driving decision selection device may be a chip or a vehicle or the like.

In an eighth aspect, an embodiment of the present application provides a computer-readable storage medium, including instructions, which, when run on a computer, cause the computer to execute the first aspect, any optional implementation manner of the first aspect, and the third aspect or the method in any optional embodiment of the third aspect.

In a ninth aspect, the embodiments of the present application provide a computer program product containing instructions, which, when run on a computer, enables the computer to execute the first aspect, any optional implementation manner of the first aspect, the third aspect or the third aspect. The method in any optional embodiment of the three aspects.

Description of drawings

1 is a schematic structural diagram of a vehicle according to an embodiment of the present application;

2 is a schematic flowchart of a driving decision selection method provided by an embodiment of the present application;

3A is a schematic diagram of a scenario of planning a driving path according to an embodiment of the application;

3B is a schematic diagram of another scenario of planning a driving path in an embodiment of the present application;

3C is a schematic diagram of another scenario of planning a driving path in an embodiment of the present application;

4A is a schematic diagram of a range interval in an embodiment of the present application;

4B is a schematic diagram of another range interval in an embodiment of the present application;

5 is a schematic diagram of another range interval in the embodiment of the application;

6A is a schematic diagram of a parking scene in an embodiment of the present application;

6B is a schematic diagram of a cockpit in another parking scene according to an embodiment of the present application;

6C is a schematic diagram of a display interface in another parking scene according to an embodiment of the present application;

6D is a schematic diagram of another parking scene in an embodiment of the present application;

FIG. 7 is a schematic structural diagram of a driving decision selection device provided by an embodiment of the present application;

FIG. 8 is a schematic structural diagram of another driving decision selection device provided by an embodiment of the present application;

FIG. 9 is a schematic structural diagram of another driving decision selection device provided by an embodiment of the present application;

FIG. 10 is a schematic structural diagram of a chip according to an embodiment of the present application.

detailed description

The technical solutions in the embodiments of the present application will be described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the present application.

The driving decision selection method provided in the embodiment of the present application can be applied to various scenarios of planning a route. Exemplarily, the present application may be applied to a scenario of selecting a driving decision for a vehicle, or the driving decision selection method provided by the present application may be performed by a vehicle. In addition, the present application can also be applied to scenarios of path planning for various types of robots, such as cargo robots, detection robots, sweeping robots or other types of robots. When carrying out transportation, it is necessary to plan a path for the freight robot to complete the transportation safely and stably.

The embodiments of the present application will be described below with reference to the accompanying drawings. Those of ordinary skill in the art know that with the development of technology and the emergence of new scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

In order to facilitate understanding of the solution, in the embodiment of the present application, the structure of the vehicle provided by the present application is first introduced with reference to FIG. 1 . Please refer to FIG. 1 first. FIG. 1 is a schematic structural diagram of a vehicle according to an embodiment of the present application. The vehicle 100 may be configured in an automatic driving mode. For example, the vehicle 100 can control itself while in the autonomous driving mode, and can confirm the current state of the vehicle and its surrounding environment through human operation, determine whether there are obstacles in the surrounding environment, and control the vehicle based on the information of the obstacles 100. The vehicle 100 may also be placed to operate without human interaction when the vehicle 100 is in an autonomous driving mode.

Vehicle 100 may include various subsystems, such as travel system 102 , sensor system 104 , control system 106 , one or more peripherals 108 and power supply 110 , computer system 112 , and user interface 116 . Alternatively, vehicle 100 may include more or fewer subsystems, and each subsystem may include multiple components. Additionally, each of the subsystems and components of the vehicle 100 may be wired or wirelessly interconnected.

The travel system 102 may include components that provide powered motion for the vehicle 100 . In one embodiment, travel system 102 may include engine 118 , energy source 119 , transmission 120 , and wheels/tires 121 .

The engine 118 may be an internal combustion engine, an electric motor, an air compression engine, or other types of engine combinations, such as a hybrid engine composed of a gasoline engine and an electric motor, and a hybrid engine composed of an internal combustion engine and an air compression engine. Engine 118 converts energy source 119 into mechanical energy. Examples of energy sources 119 include gasoline, diesel, other petroleum-based fuels, propane, other compressed gas-based fuels, ethanol, solar panels, batteries, and other sources of electricity. The energy source 119 may also provide energy to other systems of the vehicle 100 . Transmission 120 may transmit mechanical power from engine 118 to wheels 121 . Transmission 120 may include a gearbox, a differential, and a driveshaft. In one embodiment, transmission 120 may also include other devices, such as clutches. Among other things, the drive shaft may include one or more axles that may be coupled to one or more wheels 121 .

The sensor system 104 may include several sensors that sense information about the environment surrounding the vehicle 100 . For example, the sensor system 104 may include a positioning system 122 (the positioning system may be a global positioning GPS system, a Beidou system or other positioning systems), an inertial measurement unit (IMU) 124, a radar 126, a laser rangefinder 128 and camera 130. The sensor system 104 may also include sensors of the internal systems of the vehicle 100 being monitored (eg, an in-vehicle air quality monitor, a fuel gauge, an oil temperature gauge, etc.). Sensing data from one or more of these sensors can be used to detect objects and their corresponding properties (position, shape, orientation, velocity, etc.). This detection and identification is a critical function for the safe operation of the autonomous vehicle 100 . The sensors mentioned in the following embodiments of the present application may be the radar 126 , the laser rangefinder 128 or the camera 130 or the like.

Among others, the positioning system 122 may be used to estimate the geographic location of the vehicle 100 . The IMU 124 is used to sense position and orientation changes of the vehicle 100 based on inertial acceleration. In one embodiment, IMU 124 may be a combination of an accelerometer and a gyroscope. The radar 126 can use radio signals to perceive objects in the surrounding environment of the vehicle 100 , and can specifically be expressed as a millimeter-wave radar or a lidar. In some embodiments, in addition to sensing objects, radar 126 may be used to sense the speed and/or heading of objects. The laser rangefinder 128 may utilize the laser light to sense objects in the environment in which the vehicle 100 is located. In some embodiments, the laser rangefinder 128 may include one or more laser sources, laser scanners, and one or more detectors, among other system components. Camera 130 may be used to capture multiple images of the surrounding environment of vehicle 100 . Camera 130 may be a still camera or a video camera.

The control system 106 controls the operation of the vehicle 100 and its components. Control system 106 may include various components including steering system 132 , throttle 134 , braking unit 136 , computer vision system 140 , line control system 142 , and obstacle avoidance system 144 .

Among other things, the steering system 132 is operable to adjust the heading of the vehicle 100 . For example, in one embodiment it may be a steering wheel system. The throttle 134 is used to control the operating speed of the engine 118 and thus the speed of the vehicle 100 . The braking unit 136 is used to control the deceleration of the vehicle 100 . The braking unit 136 may use friction to slow the wheels 121 . In other embodiments, the braking unit 136 may convert the kinetic energy of the wheels 121 into electrical current. The braking unit 136 may also take other forms to slow the wheels 121 to control the speed of the vehicle 100 . Computer vision system 140 may be operable to process and analyze images captured by camera 130 in order to identify objects and/or features in the environment surrounding vehicle 100 . The objects and/or features may include traffic signals, road boundaries and obstacles. Computer vision system 140 may use object recognition algorithms, Structure from Motion (SFM) algorithms, video tracking, and other computer vision techniques. In some embodiments, the computer vision system 140 may be used to map the environment, track objects, estimate the speed of objects, and the like. The route control system 142 is used to plan the travel route and travel speed of the vehicle 100 . In some embodiments, the route control system 142 may include a lateral planning module 1421 and a longitudinal planning module 1422, respectively, for combining information from the obstacle avoidance system 144, the GPS 122, and one or more predetermined maps The data for the vehicle 100 plans the driving route and driving speed. Obstacle avoidance system 144 is used to identify, evaluate, and avoid or otherwise traverse obstacles in the environment of vehicle 100 , which may be embodied as actual obstacles and virtual moving bodies that may collide with vehicle 100 . In one example, the control system 106 may additionally or alternatively include components in addition to those shown and described. Alternatively, some of the components shown above may be reduced.

Vehicle 100 interacts with external sensors, other vehicles, other computer systems, or users through peripheral devices 108 . Peripherals 108 may include a wireless communication system 146 , an onboard computer 148 , a microphone 150 and/or a speaker 152 . In some embodiments, peripherals 108 provide a means for a user of vehicle 100 to interact with user interface 116 . For example, the onboard computer 148 may provide information to the user of the vehicle 100 . User interface 116 may also operate on-board computer 148 to receive user input. The onboard computer 148 can be operated via a touch screen. In other cases, peripheral devices 108 may provide a means for vehicle 100 to communicate with other devices located within the vehicle. For example, microphone 150 may receive audio (eg, voice commands or other audio input) from a user of vehicle 100 . Similarly, speakers 152 may output audio to a user of vehicle 100 . Wireless communication system 146 may wirelessly communicate with one or more devices, either directly or via a communication network. For example, wireless communication system 146 may use 3G cellular communications, such as code division multiple access (CDMA), EVDO, global system for mobile communication (GSM)/general packet radio service, GPRS), or 4G cellular communications such as LTE. Or the fifth generation mobile communication technology (5th-Generation, 5G) communication. The wireless communication system 146 may communicate using a wireless local area network (WLAN). In some embodiments, the wireless communication system 146 may communicate directly with the device using an infrared link, Bluetooth, or ZigBee. Other wireless protocols, such as various vehicle communication systems, for example, wireless communication system 146 may include one or more dedicated short range communications (DSRC) devices, which may include communication between vehicles and/or roadside stations public and/or private data communications.

The power supply 110 may provide power to various components of the vehicle 100 . In one embodiment, the power source 110 may be a rechargeable lithium-ion or lead-acid battery. One or more battery packs of such a battery may be configured as a power source to provide power to various components of the vehicle 100 . In some embodiments, power source 110 and energy source 119 may be implemented together, such as in some all-electric vehicles.

Some or all of the functions of the vehicle 100 are controlled by the computer system 112 . Computer system 112 may include at least one processor 113 that executes instructions 115 stored in a non-transitory computer-readable medium such as memory 114 . Computer system 112 may also be multiple computing devices that control individual components or subsystems of vehicle 100 in a distributed fashion. The processor 113 may be any conventional processor, such as a commercially available central processing unit (CPU). Alternatively, the processor 113 may be a dedicated device such as an application specific integrated circuit (ASIC) or other hardware-based processor. Although FIG. 1 functionally illustrates the processor, memory, and other components of the computer system 112 in the same block, one of ordinary skill in the art will understand that the processor, or memory, may actually include not stored in the same Multiple processors, or memories, within a physical enclosure. For example, memory 114 may be a hard drive or other storage medium located within a different enclosure than computer system 112 . Accordingly, references to processor 113 or memory 114 will be understood to include references to sets of processors or memories that may or may not operate in parallel. Rather than using a single processor to perform the steps described herein, some components such as the steering and deceleration components may each have their own processor that only performs computations related to component-specific functions .

In various aspects described herein, the processor 113 may be located remotely from the vehicle 100 and communicate wirelessly with the vehicle 100 . In other aspects, some of the processes described herein are performed on a processor 113 disposed within the vehicle 100 while others are performed by a remote processor 113, including taking the necessary steps to perform a single maneuver.

In some embodiments, the memory 114 may contain instructions 115 (eg, program logic) executable by the processor 113 to perform various functions of the vehicle 100 , including those described above. Memory 114 may also contain additional instructions, including instructions to send data to, receive data from, interact with, and/or control one or more of travel system 102 , sensor system 104 , control system 106 , and peripherals 108 . instruction. In addition to instructions 115, memory 114 may store data such as road maps, route information, vehicle location, direction, speed, and other such vehicle data, among other information. Such information may be used by the vehicle 100 and the computer system 112 during operation of the vehicle 100 in autonomous, semi-autonomous and/or manual modes. A user interface 116 for providing information to or receiving information from a user of the vehicle 100 . Optionally, the user interface 116 may include one or more input/output devices within the set of peripheral devices 108, such as a wireless communication system 146, an onboard computer 148, a microphone 150 or a speaker 152, and the like.

Computer system 112 may control functions of vehicle 100 based on input received from various subsystems (eg, travel system 102 , sensor system 104 , and control system 106 ) and from user interface 116 . For example, the computer system 112 may communicate with other systems or components within the vehicle 100 using a can bus, such as the computer system 112 may utilize input from the control system 106 to control the steering system 132 to avoid interference by the sensor system 104 and the obstacle avoidance system 144 Obstacles detected. In some embodiments, computer system 112 is operable to provide control of various aspects of vehicle 100 and its subsystems.

Alternatively, one or more of these components described above may be installed or associated with the vehicle 100 separately. For example, memory 114 may exist partially or completely separate from vehicle 100 . The above-described components may be communicatively coupled together in a wired and/or wireless manner.

Optionally, the above component is just an example. In practical applications, components in each of the above modules may be added or deleted according to actual needs, and FIG. 1 should not be construed as a limitation on the embodiments of the present application. The driving decision selection method provided by the present application may be executed by the computer system 112 , the radar 126 , the laser rangefinder 130 or peripheral devices such as the on-board computer 148 or other on-board terminals. For example, the driving decision selection method provided by the present application can be executed by the on-board computer 148. The on-board computer 148 can select the driving decision and plan the driving path for the vehicle, and generate control instructions according to the driving path, and send the control instructions to the computer system 112 by The computer system 112 controls the steering system 132 , the accelerator 134 , the braking unit 136 , the computer vision system 140 , the line control system 142 or the obstacle avoidance system 144 in the control system 106 of the vehicle, thereby realizing the automatic driving of the vehicle.

The above-mentioned vehicle 100 can be a car, a truck, a motorcycle, a bus, a boat, an airplane, a helicopter, a lawn mower, a recreational vehicle, a playground vehicle, construction equipment, a tram, a golf cart, a train, a cart, etc. The application examples are not particularly limited.

Decision-making and planning are an important component of autonomous driving technology. At this stage, the mainstream technical framework of the automatic driving system is shown in the figure below. Generally, the upstream perception module outputs the environment and the position information of some key objects. After the decision-making and planning module, the control target for the vehicle is generated, such as the reference for controlling the vehicle. The path or reference trajectory, etc., is output to the downstream control link to complete the closed-loop control. In the decision-making and planning process of the decision-making and planning module, it usually includes a process of estimating its own motion trajectory, and may also include a process of estimating the motion trajectories of other targets. The decision and planning module, together with the control module, usually realizes the control of the vehicle under the premise of safety, stability, speed and accuracy.

Usually, in the process of planning the driving path of the vehicle, the information collected by the sensor can be processed to obtain the perception information, including the information of the objects around the vehicle, such as the position and size of the objects within the perception range of the vehicle or the relationship with the vehicle. distance, etc., and then plan the driving path of the vehicle according to the perception information. The perception information may generally include deterministic perception information or probabilistic perception information. Deterministic perceptual information is perceptual information with a confidence level greater than a threshold. For example, perceptual information with a confidence level greater than 95% is deterministic perceptual information, and probabilistic perceptual information includes perceptual information and confidence level information.

Usually, the confidence of the perception information of an obstacle is related to the distance between the sensor and the obstacle, the environment, the size of the obstacle, etc. For example, the farther the distance between the sensor and the obstacle, the lower the confidence of the perception information of the obstacle. The closer the distance between the sensor and the obstacle, the lower the confidence of the perception information of the obstacle. Therefore, if only deterministic sensing information is used, the sensing range corresponding to the deterministic sensing information may be limited, resulting in unstable planned driving paths, and it is impossible to plan ahead for longer distance roadblocks, reducing user experience.

Therefore, the present application provides a driving decision selection method, which is used to provide a farther perception range for planning a driving path and improve user experience. The driving decision selection method provided by the present application will be described in detail below.

Please refer to FIG. 2 , a schematic flowchart of a driving decision selection method provided by the present application, as described below.

201. Obtain at least one range interval.

Wherein, the at least one range interval is obtained by dividing the detection range of the sensor based on the output information of the sensor, and the output information of the sensor may include but not limited to the distance result and distance result of the target (for ease of understanding, hereinafter referred to as obstacles). One or more of the corresponding confidence or the relationship between the distance result and the corresponding confidence, etc. The distance result may be the distance between the obstacle and the vehicle detected by the sensor, and the confidence level is used to indicate the accuracy of the distance. The distance range corresponding to the at least one range interval covers the detection range of the sensor. For example, if the detection range of the sensor covers a range with a radius of 200 meters centered on the sensor, the detection range includes multiple distance ranges, such as 0-50 meters as a distance range, 50-150 as a distance range, 150- 200 is a distance range. In addition, in the division method in the present application, the critical value of the two range intervals may be divided into the former range interval or the latter range interval, which is not limited in this application.

Optionally, each range interval also corresponds to a confidence range. Generally, there is a mapping relationship between the confidence range corresponding to each range interval and the distance range, and the mapping relationship may be the relationship between the distance result included in the output information of the sensor and the confidence level corresponding to the distance result. Degrees indicate the accuracy of the distance results detected by the sensor, and vehicle-to-object distance is the vehicle-to-object distance detected by the sensor. It can be understood that the confidence level corresponding to the at least one range interval covers the confidence level of the information detected by the sensor within the detection range. Usually, the sensor is provided in the vehicle, so the distance between the vehicle and the object is the distance between the sensor and the object. For example, the confidence that the sensor can detect within the detection range is in the range of 0-100%, and the 0-100% range is divided into multiple confidence ranges. For example, 0-80% is a confidence range, 80%-95 % is a confidence range, and more than 95% is a confidence range.

The sensor may be a sensor installed inside the vehicle, such as the sensor in the aforementioned sensor system 104, or a sensor installed outside the vehicle, such as a sensor connected to the vehicle and installed on the surface of the vehicle body. Adjust the application scene.

Specifically, when dividing the range interval, it can be divided in various manners, and several feasible division manners are exemplarily introduced below.

Method 1: Divide based on confidence

In a possible implementation, the division may be based on confidence. One or more confidence ranges are obtained for the range of confidence levels of objects detected within the detection range of the sensor. That is, the one or more confidence ranges cover the confidence of the information detected by the sensor within the detection range. Then, based on the relationship between the distance result included in the output information of the sensor and the confidence level corresponding to the distance result, the distance range corresponding to each confidence level range is calculated. Each range interval has a corresponding confidence range and a distance range, and at least one distance range corresponding to the at least one range interval covers the detection range of the sensor.

Exemplarily, as shown in Table 1, 95%-100% can be divided into range interval 1, 85%-95% can be divided into range interval 2, and so on. Moreover, each range interval has a corresponding distance range. Usually, the distance corresponding to each range interval may be different due to the error of the sensor. For example, the confidence range corresponding to 80 meters may be 93%-96%. Therefore, there may be some distance overlap between range 1 and range 2. When selecting driving decisions, vehicles can be selected based on the confidence range that does not overlap. driving decisions.

区间interval	置信度范围confidence range	距离范围(米)Distance range (meters)
范围区间1range interval 1	(95％，100％](95%, 100%]	(0，90](0,90]
范围区间2Range interval 2	(85％，95％](85%, 95%]	(70，120](70, 120]
范围区间3Range interval 3	(60％，85％](60%, 85%]	(110，170](110, 170]
范围区间4range interval 4	(40％，65％](40%, 65%]	(150，220](150, 220]
范围区间5range interval 5	[0，40％][0, 40%]	[210，+∞][210, +∞]

Table 1

The distance and confidence between the vehicle and the object monitored by the sensor may be obtained through a preset perception algorithm, or may also be referred to as an output of a perception model. The relationship between the distance and the confidence can be obtained by counting the distance between the vehicle and the object detected by the sensor, and the corresponding confidence. For ease of understanding, the relationship between the distance result included in the output information of the sensor and the confidence level corresponding to the distance result is referred to as the relationship between the distance and the confidence level below. The confidence level is the accuracy of the information of the object detected by the sensor, and the information of the detected object may include information such as the size, position, movement direction, speed, or distance to the vehicle of the object. It can be understood that the confidence can be expressed as the accuracy of information such as the size, position, movement direction, speed, or distance from the vehicle detected by the sensor.

Specifically, the relationship between the distance and the confidence may be a linear relationship, an exponential relationship, a logarithmic relationship, or a sequence of numbers, etc., which can be specifically determined according to an actual application scenario, which is not limited in this application.

In some possible scenarios, the same sensor may detect different information of an object, such as the size, direction, position and other information of the object, or the confidence of the sensor for different types of objects at the same distance may also be different, for example, if If the sensor is a camera, some content of the collected image may be unclear due to the limitation of the focus point of the camera, resulting in different confidence levels of objects at different positions. Therefore, the same sensor may have multiple distances and the relationship between the confidence of different data types, such as the relationship between the distance and the confidence of the size of the object, the relationship between the distance and the confidence of the direction of the object, etc. It can be adjusted according to actual application scenarios. For ease of understanding, in the following embodiments of the present application, only the relationship between confidence and distance of one type of data is used as an example for illustration, but not as a limitation.

In a possible implementation, the historical distance information collected by the sensor and the corresponding confidence level can be obtained, and then the distance result included in the output information of the sensor and the corresponding confidence level can be calculated according to the historical distance information and the corresponding confidence level. The relationship between the confidence levels corresponding to this distance result. For example, the information of a large number of obstacles collected by the sensor can be collected, and as the input of the preset perception algorithm, the distance between the vehicle and the object detected by the sensor and the corresponding confidence level can be output. Then, the relationship between the distance and the corresponding confidence is fitted to obtain the relationship. Exemplarily, the relationship can be expressed as y=ax, where y is the confidence, x is the distance, and a is a constant.

Further, the perception model can also be obtained by training through information collected by a large number of sensors. Exemplarily, the perception model may be one or a combination of a target detection neural network, a semantic segmentation neural network, a convolutional neural network, or a constructed network. For example, the target detection neural network may be a deep neural network for 2D target detection, such as an evolution network based on regions with CNN features (RCNN), or a deep neural network for 3D target detection, such as a Neural network of forward propagation (FP), or evolution network based on segmentation network (Segmentation Network, SegNet).

In a possible implementation manner, the relationship between the distance between the vehicle and the object detected by the sensor and the confidence level can be updated in real time through the real-time information collected by the sensor and the corresponding confidence level. For example, while the vehicle is driving, information comparing data collected by sensors at different locations can be saved to update the relationship between distance and confidence.

Method 2: Divide based on distance

In another possible implementation manner, the detection range of the sensor may be divided based on the distance to obtain one or more distance ranges.

Optionally, the confidence range corresponding to each distance range may also be calculated based on the relationship between the distance result included in the output information of the sensor and the confidence level corresponding to the distance result. The confidence range corresponding to each distance range is used for subsequent screening of range intervals that match the perception information.

For example, as shown in Table 2, the range of 0-80 can be divided into range 1, the distance of 80-120 can be divided into range 2, the distance of 120-160 can be divided into range 3, etc., and so on. Usually, due to the error of the sensor, the confidence levels corresponding to each range interval may be different. For example, the range range 1 and range range 2 may have some overlap in the confidence levels. When choosing a driving decision later, you can make a decision based on the non-overlapping distance. Scope matches perceptual information.

区间interval	距离范围(米)Distance range (meters)	置信度范围confidence range
范围区间1range interval 1	(0，80](0,80]	(98％，100％](98%, 100%]
范围区间2Range interval 2	(80，120](80, 120]	(85％,98％](85%, 98%]
范围区间3Range interval 3	(120，160](120, 160]	(62％,87％](62%, 87%]
范围区间4range interval 4	(160，200](160, 200]	(40％,65％](40%, 65%]
范围区间5range interval 5	[200，+∞][200, +∞]	[0,45％][0,45%]

Table 2

Of course, in some possible scenarios, the range interval can also be divided by distance and confidence, that is, the aforementioned method 1 and method 2 can be combined to obtain one or more scope intervals, which can be adjusted according to the actual application scenario. There are no restrictions.

Therefore, in the second method, the detection range of the sensor can be divided into one or more range intervals in combination with the confidence level, and the range interval can be used to select a driving decision, and then plan a driving path through the driving decision. Therefore, when planning a driving path, compared to using perception information with a confidence higher than a threshold to plan a driving path, the driving decision selection method provided by the present application can improve the perception range used when selecting driving decisions, and can select the driving decision of the vehicle in advance, Therefore, the vehicle can be accelerated or decelerated in advance, so that the control of the vehicle is more stable, the driving safety of the vehicle is improved, and the user experience is improved. It can be understood that the driving decision selection method provided by this application introduces probabilistic perception information to represent the uncertainty of the perception information, so that the driving decision can be selected according to the uncertain perception information in the future, which increases the planning of driving. The perception range to use when routing.

In some possible scenarios, there may be multiple sensors. When dividing the range interval, the detection range of each sensor can be divided separately to obtain the range interval corresponding to each sensor; or the same confidence level for each sensor can be obtained. The corresponding distance is divided after weighted operation, or the confidence of each sensor at the same distance is weighted and divided, etc., which can be adjusted according to the actual application scenario. Therefore, in the embodiments of the present application, different division methods can be used for different sensors, and in the process of selecting the driving decision of the vehicle, multiple sensors can work together to further improve the driving safety of the vehicle.

For example, within the range of 70m, lidar has a high confidence in the perception of vehicles or pedestrians, while in the range of 70-150m, the confidence in perception of vehicles or pedestrians is low (such as between 50% and 80%). In the range of 80m, the visual perception sensor has a high degree of confidence in the perception of vehicles or pedestrians, while in the range of 80-150m, the confidence in the perception of vehicles or pedestrians is low. Generally, by using multiple sensors to perceive the information of objects in the environment, the accuracy of detection can be improved, false detection or missed detection can be avoided, and the accuracy of the detected information can be improved.

202. Set at least one driving decision for each range section in the at least one range section.

Among them, the driving decision may include decisions such as acceleration, deceleration, cruise control, following the vehicle, maintaining the vehicle speed, or changing lanes. The driving decisions set in different ranges can be the same or different. For example, the driving decision corresponding to range 1 may be a decision such as cruising at a constant speed, slowing down and following a car, changing lanes, or speeding up, or it may also include controlling the speed of the vehicle within the first preset range or controlling the speed of the vehicle relative to the obstacle within Decisions such as the second preset range.

In some possible scenarios, the driving decision for the vehicle may be pre-set or selected by the user. For example, different distances can be preset to correspond to different decisions. The range of 0-20 meters can be set to decelerate or maintain the speed, or the corresponding speed range or relative speed range can be set. For 40-80 meters, acceleration and deceleration can be set. Or change lanes, etc., or set the corresponding speed range or relative speed range for decision-making. Above 80 meters, you can set acceleration, cruise control, deceleration, or lane change, or set the corresponding speed range or relative speed range. Decision-making. For another example, the driving decision may be selected by the user, wherein acceleration and deceleration may be set as mandatory decisions, and decisions such as cruise control and lane change may be selected by the user through an interactive interface.

It should be noted that, after setting at least one driving decision corresponding to each range in the at least one range, the vehicle can be planned by at least one divided range and the corresponding driving decision before or during driving. the driving path, and control the vehicle to drive according to the driving path. In addition, before or during the driving of the vehicle, the divided range sections or the driving decision corresponding to each range section may also be updated. For example, the sensor may have loss in the environment, resulting in errors or baseline drift, etc. Therefore, the range interval can also be updated in real time according to the information collected by the sensor during use, so as to improve the divided range interval and the accuracy of the sensor detection. degree match.

Therefore, in the embodiments of the present application, different range sections have corresponding driving decisions, so that when planning a driving route in the future, the selected driving decisions can be used for planning. It can be understood that the present application provides a segmented driving planning interval, which can process perception information with different confidence levels based on different range intervals, thereby improving the perception range when planning a driving path.

Usually, in different scenarios, the driving decision corresponding to each range interval is determined according to the application scenario. For example, the application scenario may include but not limited to scenarios such as automatic cruise, car following, or automatic parking. Specifically, for example, in a high-speed automatic cruise scenario, where the speed of the vehicle is relatively high, the distance ranges corresponding to multiple ranges may be 0m-100m, 100m-200m, 200m-300m, etc. The driving decisions corresponding to 0m-100m include deceleration or When changing lanes, the driving decision corresponding to 100m-200m is to maintain the vehicle speed or change lanes, and the driving decision corresponding to 200m-300m can include driving decisions such as acceleration and lane change. In the low-speed following scenario, the decision corresponding to 0m-100m is deceleration, and the driving decision corresponding to the distance range above 100m is acceleration. Of course, it is also possible to adapt to different scenarios by dividing different distance ranges and setting corresponding driving decisions for each distance range. For example, in a low-speed car-following scenario, the distance range corresponding to each range section may be different from that of high-speed automatic cruise. For example, the distance ranges corresponding to multiple range sections may be 0-50m, 50-100m, and 100m, respectively. -150m, the driving decisions corresponding to 0-50m are decisions such as decelerating or changing lanes, the driving decisions corresponding to 50-100m are decisions such as accelerating or maintaining the vehicle speed, and the driving decisions corresponding to 100m-150m are decisions such as accelerating or maintaining the vehicle speed, above 150m The distance range corresponds to driving decisions such as acceleration or lane change. Also, for example, in an automatic parking scenario, because the distance between the vehicle and the obstacle is relatively short, the divided distance ranges may also be different. For example, the distance ranges corresponding to multiple range intervals may include 0-1m, 1m-2m , 2m-5m, etc. The driving decision corresponding to 0-1m is to decelerate, the driving decision corresponding to 1m-2m is to maintain the vehicle speed, and the driving decision corresponding to 2m-5m is to accelerate slowly.

In addition, in different environments, the driving decisions corresponding to each range section may also be different. For example, the driving decisions corresponding to each range in a smoky environment and a non-smoky environment are different, and the driving decisions corresponding to each range in a rainy and foggy environment and in a sunny environment are also different. The driving decision can be adjusted according to the actual application scenario. This application does not limit this.

It should be noted that steps 201-202 are optional steps. The steps 201-202 can also be implemented independently, or other devices can perform the steps 201-202, and then configure at least one range interval obtained by division in the sensor, or send the at least one range interval to the vehicle terminal or other control in the equipment of the vehicle. For example, before a sensor leaves the factory, a storage medium can be set in the sensor, and an instruction code can be written in the storage medium. details of the range interval, etc.

Generally, in practical applications, it is not necessary to divide the range interval every time a driving decision is selected. , which can be adjusted according to the actual application scenario. Alternatively, in some scenarios, one or more range intervals may be demarcated in advance, and each time a driving decision is selected or a driving route is planned, the range interval does not need to be demarcated again.

203. Acquire the perception information collected by the sensor.

The vehicle is equipped with a sensor, and the sensing information collected by the sensor can be acquired during the driving of the vehicle or before starting to drive. For example, the sensor collects information such as the image of the obstacle, the laser point cloud or the electromagnetic echo point cloud, and then the information is input into a preset perception algorithm or called a perception model, and the perception information is output.

Specifically, the perception information may include, but is not limited to, one or more of the following information: the speed of the vehicle, the speed of the vehicle relative to the obstacle, the position of the obstacle, the direction of the obstacle, the size of the obstacle, and other information. The obstacles can be vehicles, pedestrians, roads, traffic lights, traffic signs, etc. within the detection range of the sensor.

Usually, the information collected by different sensors may be different. Specifically, for example, the sensor may include a camera, and the information collected by the sensor may include pixel values of the image; the sensor may include lidar, and the information collected by the sensor may include a 3D laser point cloud; the sensor may also include millimeter wave radar , the information collected by the sensor may include a point cloud composed of electromagnetic echoes. The specific type of the sensor can be adjusted according to the actual application. The information collected by different types of sensors may be the same or different. It is an exemplary illustration, not a limitation.

Different sensors may correspond to different perception models, or may correspond to the same perception model, that is, the perception models can process different types of input data to obtain corresponding perception information. The perception model can be obtained by training a large amount of historical data related to the aforementioned sensors.

Optionally, the perception information may further include a confidence level, which is used to indicate the accuracy of the information included in the perception information. For example, if the perception information can include the speed of the vehicle relative to the obstacle, the confidence level can be used to represent the accuracy of the speed of the vehicle relative to the obstacle. In order to make the subsequent selection of the driving decision of the vehicle and the planning of the driving path of the vehicle, the confidence level can be referred to for planning, and a more accurate driving path can be obtained.

204. Select a first range interval that matches the perception information from at least one range interval.

Wherein, after the sensing information is obtained through the sensing algorithm, a first range interval matching the sensing information may be selected from at least one range interval. The at least one range interval may be at least one range interval divided in the foregoing step 202, each range interval corresponds to a distance range, and optionally, each range interval also corresponds to a confidence range. For details, please refer to the foregoing step 202. description, which will not be repeated here.

In a possible implementation, the perception information may include the first distance between the vehicle and the obstacle, and step 204 may specifically include: comparing the first distance with the distance range of each range interval, and filtering out the first distance between the vehicle and the obstacle. a range interval, and the first distance is within a distance range corresponding to the first range interval. The perception information can include the distance between the vehicle and the obstacle, and the obstacle is in the driving direction of the vehicle. If it is in the same lane or the obstacle is in front of the driving direction of the vehicle, the distance between the vehicle and the obstacle can be selected. The matched range is used as the first range.

For example, if the distance between the vehicle and the obstacle included in the perception information is 90 meters, the distance range of range 1 is 50-150 meters, and the distance range of range 2 is 0-50 meters, and the range of the distance included in the perception information is 90 meters. Compared with the range interval, the 90 meters is within the range of 50-150 meters included in the range interval 1, and the range interval matching the 90 is determined as the range interval 1.

In a possible implementation manner, the perception information may include a first confidence level, where the first confidence level is used to indicate the accuracy of the distance between the vehicle and the obstacle included in the perception information. Step 204 may specifically include: matching the first confidence level with a confidence range included in each range interval, and filtering out a first range interval, where the first confidence level is within the confidence range included in the first range interval Inside.

For example, if the first confidence level is 95%, range interval 1 includes a confidence range of 94%-96%, range interval 2 includes a confidence range of 96%-100%, and 95% is included in range interval 1. In the confidence range of , that is, the range interval that matches the first confidence degree is determined as range interval 1.

Of course, in a possible implementation manner, the aforementioned at least one range interval may be divided into a combination of confidence and distance, and the perception information may include both the first distance between the vehicle and the obstacle and the first confidence of the first distance. degrees, and the obstacle is in the direction of travel of the vehicle. Then, the first range interval can be selected from the at least one range interval in combination with the first distance between the vehicle and the obstacle and the first confidence level of the distance.

For example, if the distance between the vehicle and the obstacle is 90 meters, the confidence level is 95%, the distance range corresponding to the first range interval is 50-150 meters, and the confidence level is 94%-96%, that is, the perception information includes The distance of the first range interval is within the distance range of the first range interval, and the confidence level included in the perception information is within the confidence level range corresponding to the first range range, then confirm that the screened range range is the first range range, and the first range range For the distance range of 50-150 meters, the confidence range is 94%-96%.

For another example, in a scenario, the priority of confidence and distance can be set in advance. When the confidence is in range 1 and the distance is in range 2, if the priority of the distance is higher than the priority of the confidence, then It can be confirmed that the range interval corresponding to the obstacle is the range interval 2; if the priority of the confidence is higher than the priority of the distance, it can be confirmed that the range interval corresponding to the obstacle is the range interval 1.

In addition, in some embodiments, only the detection range of the sensor may be divided to obtain a distance range corresponding to each range interval. In this scenario, the perception information can include the confidence and the relationship between the distance and the confidence, then according to the relationship between the distance and the confidence, the distance corresponding to the confidence is calculated, and then the distance and each The distance range corresponding to the range interval is matched, so as to filter out the first range interval that matches the sensing information.

In a possible scenario, there may be information detected by multiple sensors, so that multiple sensing information may be obtained, and the information and confidence levels included in each sensing information may be different. In this scenario, the one with the highest confidence among the multiple sensing information can be selected as the final sensing information, and the first range interval can be selected according to the final sensing information, or a weighted operation can be performed on the multiple sensing information, and the confidence is high. The weight value corresponding to the perceptual information is also high, so as to obtain the final perceptual information. For example, if the perception model outputs the perception information corresponding to 3 sensors, including 3 types of distances to obstacles, the weighted calculation can be performed on the 3 types of distances to obstacles, and the weight value corresponding to the distance with high confidence is also The higher the value, the distance after the weighting operation is obtained, and the first range interval is selected according to the distance after the weighting operation.

205. Select a driving decision of the vehicle from at least one driving decision in the first range interval in combination with the speed of the vehicle.

Specifically, the driving decision of the vehicle can be selected according to the preset rules in combination with the speed of the vehicle. For example, set the maximum speed and minimum speed for each range interval, if the distance between the vehicle and the obstacle is within the distance range of the first range interval, and the vehicle speed is less than the maximum speed of the first range interval, and greater than the minimum speed , the driving decision in the first range interval includes acceleration, following and deceleration, then acceleration or following can be selected as the driving decision of the vehicle; if the speed of the vehicle is less than the minimum speed, acceleration can be selected as the driving decision of the vehicle at this time, The vehicle speed is maintained between the maximum vehicle speed and the minimum vehicle speed in the first range interval.

Optionally, a maximum relative vehicle speed and a minimum relative vehicle speed may be set for each range interval. According to the speed of the vehicle and the speed of the obstacle included in the perception information, the relative speed of the vehicle to the obstacle is calculated. A driving decision for the vehicle is then selected from at least one driving decision in the first range interval based on the relative vehicle speed. For example, the driving decision in the first range interval includes acceleration, following and deceleration, if the distance between the vehicle and the obstacle is within the distance range of the first range interval, and the relative speed between the vehicle and the obstacle is less than the first range If the highest relative speed in the interval is less than the lowest relative speed, you can choose to accelerate or follow the vehicle as the vehicle's driving decision; if the relative speed between the vehicle and the obstacle is less than the lowest relative speed, the acceleration is used as the vehicle's driving decision; if the vehicle If the relative speed to the obstacle is greater than the maximum relative speed, the deceleration will be taken as the driving decision of the vehicle.

In a possible implementation, a driving decision of the vehicle may be selected from at least one driving decision corresponding to the first range interval in combination with the speed of the vehicle and the relative speed of the vehicle and the obstacle. For example, the maximum vehicle speed, minimum vehicle speed, maximum relative vehicle speed and minimum relative vehicle speed can be set for each range interval, and the driving decision of the vehicle can be selected based on the speed of the vehicle and the relative speed between the vehicle and the obstacle to keep the vehicle speed at Between the maximum vehicle speed and the minimum vehicle speed, the relative speed between the vehicle and the obstacle remains between the maximum relative vehicle speed and the minimum relative vehicle speed.

Exemplarily, taking a certain range section as an example, the distance range of the range section is 10-20 meters, and the corresponding driving decision may include decelerating and maintaining the vehicle speed. The speed of the vehicle is 30km/h, and the relative speed of the vehicle relative to the preceding vehicle is 5km/h. In order to improve the driving safety of the vehicle, the control objective is to keep the relative speed of the vehicle relative to the preceding vehicle at 0km/h. Therefore, this The driving decision at this time is to decelerate so that the relative speed of the vehicle relative to the preceding vehicle is kept at 0 km/h. Alternatively, the control target may be set at a speed of 20 km/h, and the driving decision at this time is to decelerate, thereby reducing the speed of the vehicle to 20 km/h or lower.

In addition, in some scenarios, perception information with a confidence level higher than a threshold can be processed as deterministic perception information, and the vehicle's driving decision can be selected based on the deterministic perception information, that is, a planning algorithm corresponding to the deterministic perception information can be used, such as, The compatibility of the driving decision selection method provided by the present application in planning a driving path is improved.

In some scenarios, in addition to combining the speed of the vehicle or the relative speed between the vehicle and the obstacle, other data can also be combined, such as vehicles with different driving directions from the vehicle, and the driving behavior of obstacle vehicles around the vehicle (such as lane changing, overtaking, etc. ), or changes in the driving environment, etc., to select the driving decision of the vehicle. For example, if the driving decision corresponding to the distance range of 100-200 meters is to accelerate or maintain the vehicle speed, at this time, if the weather is clear and there are no vehicles around changing lanes or overtaking, you can choose to accelerate; if the weather suddenly changes, If there is rain or fog, you can choose to maintain the speed at this time so that the vehicle can drive safely.

206. Control the vehicle to travel according to the travel decision of the vehicle.

After a driving decision for the vehicle is selected, the vehicle can be controlled to execute the driving decision. For example, if the driving decision of the vehicle is to accelerate, control the vehicle to accelerate; if the driving decision of the vehicle is to decelerate, control the vehicle to decelerate; if the driving decision of the vehicle is to maintain the speed, control the speed to remain unchanged; In order to change lanes, generate a path for the vehicle to change lanes, and control the vehicle to drive according to the path.

Specifically, the driving path of the vehicle can be planned in combination with the speed of the vehicle and the driving decision of the vehicle, and the vehicle can be controlled to travel according to the driving path. For example, if the driving decision of the vehicle is to decelerate, the driving route of the vehicle in the process of executing the deceleration decision can be generated, and the vehicle can be controlled to travel according to the driving route by controlling the steering system and braking system of the vehicle.

Generally, if the vehicle's driving decision is to accelerate, decelerate, or maintain the vehicle speed, etc., the vehicle's driving path may include a route parallel or close to the lane. If the vehicle's driving decision is to change lanes, the vehicle's driving path may include the vehicle's driving path. The curve from the current lane to the adjacent lane.

More specifically, when the driving path includes a curve, the driving path may specifically include information such as a driving curve, a steering angle, a steering radius, or a vehicle speed of the vehicle.

For ease of understanding, a specific scenario is used as an example to illustrate the method provided in this application. In a vehicle with an automatic driving function, the sensing range of a sensor in the vehicle is divided into multiple range intervals, and each range interval includes: distance range and corresponding confidence range, for example, range interval 1 includes distance range [0, 10], confidence range is [98%, 100%], range interval 2 includes distance range (10, 50 ], the confidence range is [97%, 98%), the distance range included in range interval 3 is (50, 100], the confidence range is [95%, 97%), and the distance range included in range interval 4 is (100,+ ∞], the confidence range is [0%, 95%). The corresponding driving decisions are set in advance for each range interval. For example, the driving decision in the range of [0, 10] meters includes deceleration, the driving decision in the range of (10, 50] meters includes deceleration and maintaining the vehicle speed, and the driving decision in the range of (50, 100) meters The driving decisions in the range include maintaining the speed and changing lanes, and the driving decisions in the range of (100,+∞] meters include maintaining the speed, accelerating and changing lanes. In the autonomous driving scenario, as shown in Figure 3A, during the driving of the vehicle, if It is detected that there is an obstacle 301 in front of the vehicle, and the distance between the obstacle 301 and the vehicle is 126m, which is within the distance range included in the range section 4. In the driving decision of the range section corresponding to the distance, the corresponding driving decision includes: maintaining Vehicle speed, acceleration and lane change. The driving decision can be selected according to the user's needs. For example, if the user's demand is to arrive quickly, the driving decision is to change lanes or accelerate, and if the user's demand is to drive smoothly, the driving decision is to maintain the speed, etc. Exemplarily, if the user's requirement is to arrive quickly, and the distance between the vehicle and the obstacle 301 is gradually reduced, at this time, considering the safety of the vehicle and the requirement of the user to arrive quickly, the driving decision of the vehicle is to change lanes. Then, a lane-changing driving path 302 is generated based on the current vehicle speed, the speed of obstacles and road conditions, and the vehicle is controlled to follow the driving path 302. For ease of understanding, the top view of the driving path corresponding to the lane-changing decision can be shown in FIG. 3B , after it is detected that there is an obstacle 301 in front of the vehicle, that is, in the driving direction of the vehicle, and the driving decision of the selected vehicle is to change lanes, a driving path 302 is generated, and the vehicle is controlled to travel according to the driving path. The way of generating the driving path can be Including speed-time graph (speed-time graph, ST) algorithm, 3DST algorithm (3d speed-time graph, SLT) algorithm, etc. Among them, there are many ways to plan the driving path of the vehicle, exemplarily below, with one of them The planning method is taken as an example to illustrate. As shown in FIG. 3C, the position 3031 where the vehicle is after changing lanes is selected, and two

curves

3021 and 3022 for the vehicle to travel to the position 3031 are planned, and the

curves

3021 and 3022 are tangent. Vehicle speed, calculate the turning radius of the vehicle, that is, the radius r1 of the curve 3031 and the radius r2 of the curve 3022, and then smooth the curve 3021 and the curve 3022 to obtain the driving path 302, and control the vehicle to follow the driving path 302 according to the turning radius drive.

Further, if the driving decision selection method provided by this application is performed by an on-board terminal, on-board computer or other external devices, after the on-board terminal, on-board computer or other external devices plan and obtain the driving path, the driving path can be transmitted to the aforementioned driving path. In the control system 106 shown in FIG. 1 , the control system 106 generates control commands by means of differential (PD) control, proportional, integral, and derivative (PID) control according to information such as the driving curve and vehicle speed included in the driving path, so as to The steering system 132, the accelerator 134 or the braking unit 136, etc. are controlled to control the vehicle to travel according to the driving path.

Therefore, before selecting the driving decision, at least one range interval has been divided, each range interval has a corresponding confidence level and a distance range, and there is a mapping relationship between the values included in the confidence range and the values included in the distance range. The mapping relationship is the relationship between the distance result output by the sensor and the confidence level corresponding to the distance result, and the mapping relationship may be preset or updated in real time according to the information collected by the sensor. In the process of selecting a driving decision, at least one driving decision based on each range interval can be used to select a driving decision of the vehicle, such as a decision such as acceleration, deceleration, maintaining the vehicle speed, or changing lanes. Generally, the farther away the object is detected by the sensor, the lower the confidence level. Compared with using perception information with a confidence level higher than a threshold to plan a driving route, the driving decision selection method provided by the present application can use perception information with a confidence level not higher than the threshold to select driving decisions, which is equivalent to expanding the driving route planning. range of perception. Therefore, in the driving decision selection method provided by the present application, in the process of selecting a driving decision, the driving decision can be selected by using a perceived obstacle at a longer distance, which can be understood as a driving decision can be determined for an object at a longer distance, The vehicle can avoid obstacles further away in advance, which improves the driving safety of the vehicle and improves the user experience. In addition, the obstacle that is farther away from the vehicle can be avoided in advance, and decisions such as acceleration or deceleration can be made in advance, so that the driving process of the vehicle is smoother and the user experience is improved.

Further, for ease of understanding, the following is an exemplary description of a specific application scenario for the multiple range intervals divided by the driving decision selection method provided by the present application and the selection manner of the driving decision.

Exemplarily, as shown in FIG. 4A , the sensing range of the sensor provided on the vehicle 401 can be divided into 4 sections, the distance range of section 1 is 0-20 meters (m), and the distance range of section 2 is 20-65 meters. meters, the distance range for interval 3 is 65-150 meters, and the distance range for interval 4 is 150-250 meters. The critical distance can be divided into any one of two adjacent intervals. For example, a distance of 20m can be divided into interval 1 or interval 2, which can be adjusted according to the actual application scenario.

Then set the corresponding driving decisions for each section. For example, for the automatic driving mode, the driving decisions set for section 1 include braking and keeping following, and the driving decisions set for section 2 include following, changing lanes, and decelerating. Driving decisions set for 3 include accelerating, following, changing lanes, and decelerating, and driving decisions set for zone 4 include accelerating.

Among them, the sensing areas closer to the vehicle, such as interval 1 and interval 2, within this range, the information detected by the sensor usually has a high confidence level, which can be understood as the range interval corresponding to the deterministic sensing information. In the sensing area that is far from the vehicle, such as interval 3 or interval 4, within this range, the information detected by the sensor usually has a low confidence level, that is, the accuracy of the detected information is low, which can be understood as probabilistic perception information. the corresponding range. Therefore, the range interval divided in the embodiment of the present application increases the distance range of the available interval 3 and interval 4, which is used to select the driving decision of the vehicle, thereby planning the driving path of the vehicle, and increases the available perception range when planning the driving path.

Therefore, the solution provided by the present application is equivalent to increasing the perception range when planning a driving path. For example, in the range of 100m, visual perception can detect obstacles more reliably, which is a deterministic perception range; while in the range of 100-200m, due to a series of reasons such as the distance becomes farther or the image becomes smaller, the success rate of detecting vehicles decreases, the confidence also decreases. In the solution provided in this application, the 100-200m interval is also divided into regions, and the driving decision of the vehicle is selected based on this, and the driving path of the vehicle is planned, which is equivalent to increasing the perception range when planning the driving path.

In addition, referring to FIG. 4B , taking the obstacle 402 as the preceding vehicle as an example, Vr can be understood as the relative speed between the vehicle and the preceding vehicle, and Ve can be understood as the speed of the vehicle itself. Among them, the vehicle in front can be in a moving state, that is, the speed of the vehicle in front is not 0, or it can be in a stationary state, that is, the speed of the vehicle in front is 0, or the vehicle in front here can also be replaced with other obstacles, such as traffic lights, roadblocks, triangles cones, etc.

When choosing a driving decision, if the distance between the obstacle 402 and the vehicle 401 is within the distance range of section 1, the section 1 can be understood as an emergency braking section, that is, the distance between the vehicle and the obstacle 402 is over If the distance is too close, it is not within the safe distance, and the difference from the safe distance is large, so the speed of the vehicle needs to be reduced to maintain a safer distance between the vehicle and the obstacle. At this time, the control target of the vehicle may be to stop or keep the vehicle speed within a range of 0≤Vr≤30, 0≤Ve≤60, so as to avoid a rear-end collision between the vehicle and the preceding vehicle. The safety distance can be a value set in advance, or a value calculated according to a set algorithm, such as a value calculated according to the current speed of the vehicle, the braking distance or the speed of the preceding vehicle, etc. The safety distance can be understood as guaranteeing the vehicle The distance you can travel safely to avoid a collision.

If the distance obstacle between the preceding vehicle 402 and the vehicle 401 is within the distance range of interval 2, the distance between the vehicle and the preceding vehicle is in the normal braking interval, that is, the distance between the vehicle and the preceding vehicle is not within the safe distance, but is within the safe distance. The difference between them is small, and the control target of the vehicle 401 can be set as 30≤Vr≤60, 60≤Ve≤100, at this time, the vehicle speed of the vehicle 401 can be appropriately reduced so that the vehicle speed is within the range of the control target.

If the distance obstacle between the preceding vehicle 402 and the vehicle 401 is within the distance range of Section 3, the distance between the preceding vehicle and the preceding vehicle is in the comfortable speed regulation zone, and the distance between the preceding vehicle and the vehicle is not less than the safe distance. In this scene, the speed of the vehicle can be adjusted according to the actual scene, and the control target can be set as 60≤Vr≤100, 100≤Ve≤125, so that the vehicle can keep running fast and safely.

If the distance obstacle between 402 and the vehicle 401 is within the distance range of interval 4, the distance between the vehicle and the preceding vehicle is in the pre-judgment interval, the distance between the preceding vehicle and the vehicle is not less than the safe distance, and the vehicle is in a relatively safe position. In this scene, the speed of the vehicle can be adjusted according to the actual scene, and the control target can be set as 100≤Vr≤150, 125≤Ve≤150, so that the vehicle can keep running fast and safely. In addition, obstacles within the distance range from the vehicle can be avoided in advance, thereby further improving the driving safety of the vehicle.

Of course, in addition to selecting the control target with reference to Vr and Ve, it is also possible to directly select the vehicle acceleration or deceleration driving decision based on the distance between the vehicle and the aforementioned distance as the control distance. For example, if the distance between the vehicle and the obstacle is too close, control the vehicle to decelerate, Thereby increasing the distance between the vehicle and the obstacle; if the distance between the vehicle and the obstacle is too far, control the vehicle to maintain the speed or accelerate, so as to maintain the distance or reduce the distance, which can be adjusted according to the actual application scenario, here It is only an illustration, not a limitation.

For ease of understanding, it can be understood that, among the four intervals shown in FIG. 4A, interval 1 and interval 2 are the range intervals corresponding to the deterministic perceptual information, that is, the range intervals corresponding to the perceptual information whose confidence is higher than the threshold. 3 and interval 4 are the range intervals corresponding to the perceptual information whose confidence is not higher than the threshold. When it is confirmed that the distance between the vehicle and the obstacle is in the interval 1 or interval 2 according to the confidence or distance carried in the perception information, the existing method of planning the driving path according to the selective perception information can be used to plan the driving path for the vehicle. Thus, the compatibility of the way of planning the driving path in the driving decision selection method provided by the present application is improved. When it is confirmed that the distance between the vehicle and the obstacle is in section 3 or section 4 according to the confidence level or distance carried in the perception information, the driving decision of the vehicle can be selected in combination with the speed of the vehicle and/or the relative speed between the vehicle and the obstacle, so as to determine the distance between the vehicle and the obstacle in advance. More distant obstacles are dealt with in advance, such as avoiding obstacles, decelerating in advance, etc., thereby improving the safety of vehicle driving, making the driving process of the vehicle change more smoothly, and improving the user experience.

More specifically, in order to facilitate understanding, the process of selecting the driving decision of the vehicle will be described in more detail.

Taking the selection of the control target with reference to Vr and Ve as an example, after confirming that the distance between the vehicle and the obstacle so far is within the distance range of a certain range, the speed control target of the boundary of the range can be set, and then the boundary of the range can be set. The speed control target is used as the vehicle speed control target.

Exemplarily, as shown in FIG. 5 , taking one of the range intervals, such as interval i as an example, when the perceived distance between the obstacle 402 and the vehicle is in the interval i, that is, the range memory of the interval i perceived by the vehicle. In the obstacle, take the obstacle as the front vehicle as an example.

If the relative distance between the vehicle and the preceding vehicle continues to decrease, the relative distance between the vehicle and the preceding vehicle will approach the interval i-1. At this time, Vr1=60km/h and Ve1=100km/h are the boundary vehicle speeds. The boundary vehicle speed is used as the control target to plan the deceleration motion to generate the driving path of the vehicle. At this time, the driving path of the vehicle can be to continue driving in the current lane, such as driving in a straight line. When the relative distance between the vehicle and the preceding vehicle decreases to the boundary of the section i, the control targets are Vr1=60km/h and Ve1=100km/h. If the speed of the vehicle is greater than 100km/h and the relative speed between the vehicle and the vehicle in front is greater than 60km/h, reduce the vehicle speed, that is, the driving decision is to decelerate, so that the vehicle speed is not greater than 100km/h, and the relative speed between the vehicle and the vehicle in front is not greater than 60km/h h.

If the relative distance between the vehicle and the preceding vehicle continues to increase, the relative distance between the vehicle and the preceding vehicle will approach the interval i+1. At this time, Vr2=100km/h and Ve2=125km/h are the boundary speeds for the movement of the vehicle. The trajectory and speed of the vehicle are controlled. When the distance to the preceding vehicle is reduced to the boundary of the section i, the control target is to make the own vehicle speed Ve2=125 km/h and the relative speed to the preceding vehicle Vr2=100 km/h.

Therefore, in the embodiment of the present application, the range interval is divided according to the confidence and the distance of the object detected by the sensor, and the larger range is divided. Compared with only using the perception information with higher confidence to plan the driving path, the present application The methods provided are equivalent to increasing the perception range. And select the driving decision corresponding to each range interval. The range interval corresponding to the relative distance between the vehicle and the obstacle can be known, and the corresponding driving decision can be selected according to the range interval, so that the vehicle can choose the driving decision and plan the driving path for the obstacles in the farther distance, such as decelerating in advance to avoid Collision improves the driving safety of the vehicle, and can perform smoother acceleration or deceleration in advance, which can improve the user experience.

In some scenarios, for different types of obstacles, the types of driving decisions set in each range are also different. For example, if the obstacle is a moving object, the set driving decision can refer to the introduction in the aforementioned FIG. 4A . For another example, if the obstacle is a stationary object, when the vehicle gradually approaches the obstacle, the driving decision set for each range section only includes deceleration or lane change, but not acceleration, so as to improve the driving safety of the vehicle. .

In addition, the driving decision selection method provided by the present application can also be applied to parking scenarios in addition to planning the path of the vehicle during driving. Exemplarily, a parking scenario may be shown in FIG. 6A , where a vehicle 601 is in a parking lot, and there are one or more unavailable parking spaces 602 in the parking lot, and one or more available parking spaces 603 for unparked vehicles. After the automatic parking is activated, the cockpit of the vehicle can refer to FIG. 6B, and the display screen in the dashboard of the vehicle can display that the vehicle is currently in the automatic parking mode. There can be one or more parking spaces 603, which can be displayed during the interaction of the vehicle. Available parking spaces are displayed in the display interface 1000, and the user selects a parking space to be parked in. The interface displayed by the interactive display interface 1000 may be as shown in FIG. 6C , and the user may select one of the parking spaces 6031 among the plurality of parking spaces 603 . Subsequently, the scene of automatic parking in the parking space is shown in Figure 6D. According to the distance and confidence of the obstacles in the environment detected by the sensor, the driving decision corresponding to the range of the distance between the vehicle and the obstacle can be selected, so as to The driving path 604 for the vehicle 601 to be parked in the parking space 6031 is planned. For the specific method of planning the driving path, reference may be made to the related introduction in FIG. 3C , which will not be repeated here.

The flow of the driving decision selection method provided by the present application has been described in detail above, and the device provided by the present application will be described below with reference to the method embodiments corresponding to the foregoing FIGS. 2-6D .

Please refer to FIG. 7 , which is a schematic structural diagram of a driving decision selection device provided by the present application. The driving decision selection device is used to execute the steps of the method corresponding to the aforementioned FIGS. 2-6D .

The driving decision selection device may include:

A perception module 701, configured to acquire perception information collected by sensors configured on the vehicle;

A decision-making module 702, configured to select a first range interval that matches the sensing information from at least one range interval, the at least one range interval is obtained by dividing the detection range of the sensor based on the output information of the sensor, and the output information of the sensor includes At least one of the distance result of the detected target, the confidence level corresponding to the distance result, or the relationship between the distance result and the confidence level corresponding to the distance result, each range interval in the at least one range interval corresponds to at least one driving decision, and each range interval corresponds to at least one driving decision. The range interval has corresponding at least one driving decision;

The decision-making module 702 is further configured to select a driving decision of the vehicle from at least one driving decision corresponding to the first range interval according to the speed of the vehicle;

The control module 703 is used to control and measure the driving according to the driving decision of the vehicle.

In a possible implementation manner, the decision module 702 is specifically configured to select a first range interval that matches the first distance from at least one range interval, and the first distance is within the distance range included in the first range interval.

In a possible implementation, each range interval further includes a confidence range, the confidence range corresponding to at least one range interval covers the confidence of the information detected by the sensor within the detection range, and the sensing information further includes the first confidence degree, the first confidence degree is used to represent the accuracy of the first distance;

The decision-making module 702 is specifically configured to select a first range interval that matches the first confidence level from at least one range interval, the first confidence level is included in the perception information, and the first confidence level is used to indicate the accuracy of the first distance degree.

In a possible implementation manner, the driving decision selection device provided by the present application may further include: a dividing module 705, configured to divide the detection range of the sensor before the sensing module acquires sensing information, to obtain at least one A distance range, the at least one distance range is in one-to-one correspondence with the at least one range interval.

In a possible implementation manner, the dividing module 705 is specifically configured to: before the sensing module 701 obtains the sensing information, obtain at least one confidence range, and each confidence range in the at least one confidence range does not overlap, and the at least one confidence range does not overlap. A confidence level covers the confidence level of the information detected by the sensor within the detection range; according to at least one confidence level range and the relationship between the distance result and the confidence level corresponding to the distance result, calculate and at least one confidence level The ranges correspond to at least one distance range one-to-one, and each range interval corresponds to a confidence range and a distance range.

In a possible implementation, the dividing module 705 is further configured to: determine at least one confidence range corresponding to at least one distance range one-to-one according to the relationship between the distance result and the confidence level corresponding to the distance result, at least one confidence level The range is used to filter the range interval that matches the perceptual information from at least one range interval.

In a possible implementation manner, the perception module 701 is further configured to: acquire historical distance information collected by sensors, historical distance information and corresponding confidence levels; acquire distance results and distance results corresponding to the historical distance information and corresponding confidence levels relationship between confidence levels.

In a possible implementation, the decision-making module 702 is specifically configured to: calculate the relative speed between the vehicle and the obstacle according to the speed of the vehicle; and select the relative speed of the vehicle from at least one driving decision corresponding to the first range interval in combination with the relative speed. driving decisions.

In a possible implementation, the control module 703 may control the vehicle to travel according to the driving path through the vehicle actuator 704 of the vehicle. The vehicle actuator 704 may include one or more of the aforementioned modules of the travel system 102 or the control system 106 of FIG. 1 .

In a possible implementation, at least one driving decision corresponding to each range interval is determined according to an application scenario, and the application scenario includes but is not limited to scenarios such as automatic cruise, car following, or automatic parking.

Please refer to FIG. 8 , which is a schematic structural diagram of another driving decision and selection device provided by the present application. The driving decision selection device is used to execute the steps of the method corresponding to the aforementioned FIGS. 2-6D .

A division module 801, configured to obtain at least one range interval, each range interval in the at least one range interval has a corresponding confidence range and a distance range, and the distance range included in the at least one range interval covers the detection range of the sensor, at least a confidence range covering the confidence of the information detected by the sensor within the detection range;

The decision module 802 is used to set at least one driving decision corresponding to each range interval, each range interval and at least one driving decision corresponding to each range interval are used to select the driving decision of the vehicle, and the driving decision of the vehicle is used to generate The driving path of the vehicle.

In a possible implementation manner, the dividing module 801 is specifically used for:

Divide the detection range of the sensor based on the distance between the vehicle and the object detected by the sensor to obtain at least one distance range, and calculate each A confidence range corresponding to a distance range, each range interval corresponds to a confidence range and a distance range, and a range range includes a confidence range and a distance range.

Divide the range of confidence levels that can be detected by the sensor to obtain at least one confidence level range, and calculate the corresponding confidence level according to the relationship between the distance result included in the output information of the sensor and the confidence level corresponding to the distance result. Distance ranges, each range interval corresponds to a confidence range and a distance range.

In a possible implementation, the driving decision selection device may further include: a control module 805 and a perception module 804;

A perception module 804, configured to acquire perception information, where the perception information includes information about obstacles, such as the first distance between the vehicle and the obstacle;

The control module 805 is configured to select a driving decision of the vehicle from at least one driving decision in the first range interval according to the perception information, and control the vehicle to travel according to the driving decision.

In a possible implementation manner, the perception information includes the distance between the obstacle and the vehicle, and the control module 805 is specifically used for:

It is determined that the confidence level included in the perception information is within the confidence level range included in the first range interval; and a driving decision of the vehicle is selected from at least one driving decision in the first range interval in combination with the speed of the vehicle. In a possible implementation manner, the control module 805 is further configured to:

According to the speed of the vehicle, the relative speed of the vehicle relative to the obstacle is calculated, and then, in combination with the relative speed, a driving decision of the vehicle is selected from at least one driving decision in the first range interval. In a possible implementation, the perception information further includes the speed of the vehicle and/or the relative speed of the vehicle and the obstacle;

The control module 805 is specifically configured to select a driving decision of the vehicle from at least one driving decision in the first range interval in combination with the speed of the vehicle and/or the relative speed of the vehicle and the obstacle.

In a possible implementation, the confidence level included in the perception information is related to the distance between the sensor and the obstacle.

In a possible implementation manner, the driving decision selection device may further include an acquisition module 803, which is specifically used for:

Obtain the historical distance information collected by the sensor and the corresponding confidence;

The relationship between the distance result included in the output information of the sensor and the confidence level corresponding to the distance result is obtained according to the historical distance information and the corresponding confidence level.

Optionally, the vehicle actuator 806 is similar to the aforementioned vehicle actuator 704 and will not be repeated here.

Please refer to FIG. 9 , a schematic structural diagram of another driving decision and selection device provided by the present application, as described below.

The driving decision selection device may include a processor 901 and a memory 902 . The processor 901 and the memory 902 are interconnected by wires. Among them, the memory 902 stores program instructions and data.

The memory 902 stores program instructions and data corresponding to the aforementioned steps in FIGS. 2-6D .

The processor 901 is configured to perform the method steps performed by the driving decision selection apparatus shown in any of the foregoing embodiments in FIGS. 2-6D .

Optionally, the driving decision selection device may further include a transceiver 903 for receiving or transmitting data.

Embodiments of the present application also provide a computer-readable storage medium, where a program for generating a vehicle's running speed is stored in the computer-readable storage medium, and when the computer is running on a computer, the computer is made to execute the program as shown in the foregoing FIGS. 2-6D . Steps in the method described in the example embodiment.

Optionally, the aforementioned driving decision selection device shown in FIG. 9 is a chip.

The embodiments of the present application also provide a driving decision-making selection device, which may also be referred to as a digital processing chip or a chip. The chip includes a processing unit and a communication interface. The processing unit obtains program instructions through the communication interface, and the program instructions are processed. The unit is executed, and the processing unit is configured to execute the method steps executed by the driving decision selection apparatus shown in any of the foregoing embodiments in FIGS. 2-6D .

The embodiments of the present application also provide a digital processing chip. The digital processing chip integrates circuits and one or more interfaces for realizing the above-mentioned processor 901 or the functions of the processor 901 . When a memory is integrated in the digital processing chip, the digital processing chip can perform the method steps of any one or more of the foregoing embodiments. When the digital processing chip does not integrate the memory, it can be connected with the external memory through the communication interface. The digital processing chip implements the actions performed by the driving decision and selection device in the above embodiment according to the program codes stored in the external memory.

The embodiments of the present application also provide a computer program product that, when driving on the computer, causes the computer to execute the steps performed by the driving decision selection device in the method described in the embodiments shown in FIGS. 2-6D.

The driving decision selection device provided in the embodiment of the present application may be a chip, and the chip includes: a processing unit and a communication unit, the processing unit may be, for example, a processor, and the communication unit may be, for example, an input/output interface, a pin, or a circuit, etc. . The processing unit can execute the computer-executed instructions stored in the storage unit, so that the chip executes the driving decision selection method described in the embodiments shown in FIGS. 2-6D . Optionally, the storage unit is a storage unit in the chip, such as a register, a cache, etc., and the storage unit may also be a storage unit located outside the chip in the wireless access device, such as only Read-only memory (ROM) or other types of static storage devices that can store static information and instructions, random access memory (RAM), etc.

Specifically, the aforementioned processing unit or processor may be a central processing unit (CPU), a network processor (neural-network processing unit, NPU), a graphics processing unit (graphics processing unit, GPU), a digital signal processing digital signal processor (DSP), application specific integrated circuit (ASIC) or field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general purpose processor may be a microprocessor or it may be any conventional processor or the like.

Exemplarily, please refer to FIG. 10. FIG. 10 is a schematic structural diagram of a chip provided by an embodiment of the present application. The chip may be represented as a neural network processor NPU 100, and the NPU 100 is mounted on the main CPU ( Host CPU), the task is allocated by the Host CPU. The core part of the NPU is the operation circuit 100, and the operation circuit 1003 is controlled by the controller 1004 to extract the matrix data in the memory and perform multiplication operations.

In some implementations, the arithmetic circuit 1003 includes multiple processing units (process engines, PEs). In some implementations, the arithmetic circuit 1003 is a two-dimensional systolic array. The arithmetic circuit 1003 may also be a one-dimensional systolic array or other electronic circuitry capable of performing mathematical operations such as multiplication and addition. In some implementations, arithmetic circuit 1003 is a general-purpose matrix processor.

For example, suppose there is an input matrix A, a weight matrix B, and an output matrix C. The arithmetic circuit fetches the data corresponding to the matrix B from the weight memory 1002 and buffers it on each PE in the arithmetic circuit. The arithmetic circuit fetches the data of matrix A and matrix B from the input memory 1001 to perform matrix operation, and stores the partial result or final result of the matrix in an accumulator 1008 .

Unified memory 1006 is used to store input data and output data. The weight data is directly passed through the storage unit access controller (direct memory access controller, DMAC) 1005, and the DMAC is transferred to the weight memory 1002. Input data is also transferred to unified memory 1006 via the DMAC.

A bus interface unit (BIU) 1010 is used for the interaction between the AXI bus and the DMAC and an instruction fetch buffer (instruction fetch buffer, IFB) 1009.

The bus interface unit 1010 (bus interface unit, BIU) is used for the instruction fetch memory 1009 to obtain instructions from the external memory, and is also used for the storage unit access controller 1005 to obtain the original data of the input matrix A or the weight matrix B from the external memory.

The DMAC is mainly used to transfer the input data in the external memory DDR to the unified memory 1006 , the weight data to the weight memory 1002 , or the input data to the input memory 1001 .

The vector calculation unit 1007 includes a plurality of operation processing units, and further processes the output of the operation circuit, such as vector multiplication, vector addition, exponential operation, logarithmic operation, size comparison, etc., if necessary. It is mainly used for non-convolutional/fully connected layer network computations in neural networks, such as batch normalization, pixel-level summation, and upsampling of feature planes.

In some implementations, the vector computation unit 1007 can store the vector of processed outputs to the unified memory 1006 . For example, the vector calculation unit 1007 may apply a linear function and/or a nonlinear function to the output of the operation circuit 1003, such as linear interpolation of the feature plane extracted by the convolutional layer, such as a vector of accumulated values, to generate activation values. In some implementations, the vector computation unit 1007 generates normalized values, pixel-level summed values, or both. In some implementations, the vector of processed outputs can be used as activation input to the arithmetic circuit 1003, eg, for use in subsequent layers in a neural network.

The instruction fetch memory (instruction fetch buffer) 1009 connected to the controller 1004 is used to store the instructions used by the controller 1004;

The unified memory 1006, the input memory 1001, the weight memory 1002 and the instruction fetch memory 1009 are all On-Chip memories. External memory is private to the NPU hardware architecture.

The operation of each layer in the recurrent neural network can be performed by the operation circuit 1003 or the vector calculation unit 1007 .

Wherein, the processor mentioned in any one of the above may be a general-purpose central processing unit, a microprocessor, an ASIC, or one or more integrated circuits used to control the execution of the programs of the above-mentioned methods of FIGS. 2-6D .

In addition, it should be noted that the device embodiments described above are only schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be A physical unit, which can be located in one place or distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution in this embodiment. In addition, in the drawings of the device embodiments provided in the present application, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines.

From the description of the above embodiments, those skilled in the art can clearly understand that the present application can be implemented by means of software plus necessary general-purpose hardware. Special components, etc. to achieve. Under normal circumstances, all functions completed by a computer program can be easily implemented by corresponding hardware, and the specific hardware structures used to implement the same function can also be various, such as analog circuits, digital circuits or special circuit, etc. However, a software program implementation is a better implementation in many cases for this application. Based on this understanding, the technical solutions of the present application can be embodied in the form of software products in essence, or the parts that make contributions to the prior art. The computer software products are stored in a readable storage medium, such as a floppy disk of a computer. , U disk, mobile hard disk, read only memory (ROM), random access memory (RAM), disk or CD, etc., including several instructions to make a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present application.

In the above-mentioned embodiments, it may be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented in software, it can be implemented in whole or in part in the form of a computer program product.

The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present application are generated. The computer may be a general purpose computer, special purpose computer, computer network, or other programmable device. The computer instructions may be stored in or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be downloaded from a website site, computer, server, or data center Transmission to another website site, computer, server, or data center is by wire (eg, coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (eg, infrared, wireless, microwave, etc.). The computer-readable storage medium may be any available medium that can be stored by a computer, or a data storage device such as a server, data center, etc., which includes one or more available media integrated. The usable media may be magnetic media (eg, floppy disks, hard disks, magnetic tapes), optical media (eg, DVDs), or semiconductor media (eg, solid state disks (SSDs)), and the like.

The terms "first", "second", "third", "fourth", etc. (if any) in the description and claims of this application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It is to be understood that data so used may be interchanged under appropriate circumstances so that the embodiments described herein can be practiced in sequences other than those illustrated or described herein. Furthermore, the terms "comprising" and "having" and any variations thereof, are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those expressly listed Rather, those steps or units may include other steps or units not expressly listed or inherent to these processes, methods, products or devices.

Claims

A driving decision selection method, comprising:

Obtain the perception information collected by the sensors configured on the vehicle;

A first range interval matching the sensing information is selected from at least one range interval, the at least one range interval is obtained by dividing the detection range of the sensor based on the output information of the sensor, and the output of the sensor The information includes at least one of the distance result of the detected target, the confidence level corresponding to the distance result, or the relationship between the distance result and the confidence level corresponding to the distance result, and each of the at least one range interval corresponds to at least one driving decision making;

In combination with the speed of the vehicle, a driving decision of the vehicle is selected from at least one driving decision corresponding to the first range interval, and the vehicle is controlled to travel according to the driving decision of the vehicle.
The method according to claim 1, wherein each range interval corresponds to a distance range, the perception information includes a first distance, and the first distance is a distance between the vehicle and an obstacle,

The selecting a first range interval that matches the perceptual information from at least one range interval includes:

From the at least one range interval, the first range interval that matches the first distance is selected.
The method according to claim 1, wherein each range interval corresponds to a confidence range, the perception information includes a first confidence level, and the first confidence level is used to indicate that the perception information contains the degree of accuracy with which the distance of the vehicle to the obstacle is included;

The selecting a first range interval that matches the perceptual information from at least one range interval includes:

From the at least one range interval, the first range interval that matches the first confidence level is selected, and the first confidence level is within the confidence level range included in the first range interval.
The method according to any one of claims 1-3, wherein, before the acquiring the perception information, the method further comprises:

The detection range of the sensor is divided to obtain at least one distance range, and the at least one distance range is in one-to-one correspondence with the at least one range interval.
The method according to claim 4, wherein, dividing the detection range of the sensor to obtain at least one distance range, comprising:

acquiring at least one confidence level range, each confidence level range in the at least one confidence level range does not overlap, and the at least one confidence level range covers the confidence level of the information detected by the sensor within the detection range;

According to the at least one confidence range and the relationship between the distance result and the confidence corresponding to the distance result, the detection range of the sensor is divided into at least one distance range.
The method according to claim 4, wherein after dividing the detection range of the sensor to obtain at least one distance range, the method further comprises:

According to the relationship between the distance result and the confidence level corresponding to the distance result, at least one confidence level range corresponding to the at least one distance range is determined, and the at least one confidence level range is used to select from the at least one range interval. Filter the range interval that matches the perception information.
The method according to any one of claims 1-6, wherein the method further comprises:

Obtain the historical distance information and the corresponding confidence level of the historical distance information collected by the sensor;

The relationship between the distance result and the confidence level corresponding to the distance result is obtained according to the historical distance information and the corresponding confidence level.
The method according to any one of claims 1-7, wherein the driving decision of the vehicle is selected from at least one driving decision in the first range interval in combination with the speed of the vehicle, include:

Calculate the relative speed of the vehicle and the obstacle according to the speed of the vehicle;

In combination with the relative speed, a driving decision of the vehicle is selected from at least one driving decision corresponding to the first range interval.
The method according to any one of claims 1-8, wherein the at least one driving decision corresponding to each range interval is determined according to an application scenario, and the application scenario includes: automatic cruise, car following or automatic Parking.
A driving decision selection device, characterized in that it includes:

The perception module is used to obtain the perception information collected by the sensors configured on the vehicle;

a decision-making module, configured to select a first range interval matching the sensing information from at least one range interval, the at least one range interval being obtained by dividing the detection range of the sensor based on the output information of the sensor, The output information of the sensor includes at least one of the distance result of detecting the target, the confidence level corresponding to the distance result, or the relationship between the distance result and the confidence level corresponding to the distance result, and each range in the at least one range interval The interval corresponds to at least one driving decision;

The decision-making module is further configured to select a driving decision of the vehicle from at least one driving decision corresponding to the first range interval in combination with the speed of the vehicle;

A control module, configured to control the vehicle to travel according to the travel decision of the vehicle.
The device according to claim 10, wherein each range interval corresponds to a distance range, the perception information further includes a first distance, and the first distance is a distance between the vehicle and an obstacle;

The decision-making module is specifically configured to select the first range interval that matches the first distance from the at least one range interval, and the first distance is within the distance range included in the first range interval Inside.
The apparatus according to claim 10, wherein each range interval corresponds to a confidence range, the perception information includes a first confidence level, and the first confidence level is used to indicate that the perception information contains the degree of accuracy with which the distance of the vehicle to the obstacle is included;

The decision-making module is specifically configured to, based on a first confidence level, select the first range interval with the first confidence level from the at least one range interval, and the first confidence level is included in the perception In the information, the first confidence level is used to represent the accuracy of the first distance.
The device according to any one of claims 10-12, wherein the device further comprises:

The dividing module is configured to divide the detection range of the sensor before the sensing module acquires sensing information to obtain at least one distance range, and the at least one distance range corresponds to the at least one range interval one-to-one.
The device according to claim 13, wherein each range interval includes a confidence range, and the device further comprises: a dividing module, which is specifically configured to:

acquiring at least one confidence level range, each confidence level range in the at least one confidence level range does not overlap, and the at least one confidence level range covers the confidence level of the information detected by the sensor within the detection range;

According to the at least one confidence range and the relationship between the distance result and the confidence corresponding to the distance result, at least one distance range corresponding to the at least one confidence range is obtained, and the at least one confidence range is obtained. The range and the at least one distance range make up the at least one range interval.
The apparatus according to claim 13, wherein the dividing module is further configured to determine at least one one-to-one correspondence with the at least one distance range according to the relationship between the distance result and the confidence level corresponding to the distance result A confidence range, where the at least one confidence range is used to filter a range interval that matches the perceptual information from the at least one range interval.
The device according to any one of claims 10-15, wherein the sensing module is further configured to:

Obtain the historical distance information and the corresponding confidence level of the historical distance information collected by the sensor;

The relationship between the distance result and the confidence level corresponding to the distance result is obtained according to the historical distance information and the corresponding confidence level.
The device according to any one of claims 10-16, wherein the decision-making module is specifically configured to:

Calculate the relative speed of the vehicle and the obstacle according to the speed of the vehicle;

In combination with the relative speed, a driving decision of the vehicle is selected from at least one driving decision corresponding to the first range interval.
The device according to any one of claims 10-17, wherein the at least one driving decision corresponding to each range interval is determined according to an application scenario, and the application scenario includes: automatic cruise, following a car, or automatic Parking.
A driving decision selection device, characterized in that it includes a processor;

The processor obtains program instructions through the communication interface, and when the program instructions are executed by the processing unit, the method of any one of claims 1-9 is implemented; or,

The processor is coupled to a memory, and the memory stores a program, the program instructions stored in the memory, when executed by the processor, implement the method of any one of claims 1 to 9.
A computer-readable storage medium comprising a program which, when executed by a processing unit, performs the method of any one of claims 1 to 9.