EP-4737255-A2 - CONTROL METHOD AND APPARATUS

EP4737255A2EP 4737255 A2EP4737255 A2EP 4737255A2EP-4737255-A2

Abstract

A method applied to the field of self-driving technologies is provided. The method includes: controlling a target device in a vehicle to display first request information in a target area, where the first request information is used to request a pedestrian to perform a target action, and the target action is used to express a road participation intention of the pedestrian; recognizing an action performed by the pedestrian; and determining a driving policy of the vehicle based on a recognition result. A control apparatus is also provided. Therefore, even without participation of a driver, the vehicle may query the road participation intention of the pedestrian by using the first request information, so that in a self-driving scenario, the vehicle can implement intention interaction with the pedestrian, to obtain an accurate road participation intention of the pedestrian, and obtain an appropriate driving policy. This improves driving safety.

Inventors

LIU, HANG
SUI, Linlin
XIA, Xuan

Assignees

Shenzhen Yinwang Intelligent Technologies Co., Ltd.

Dates

Publication Date: 20260506
Application Date: 20220519

Claims (15)

A control method, wherein the method comprises: controlling (S801, S1201) a target device in a vehicle to display first request information in a target area, wherein the first request information is used to request a pedestrian to perform a first target action, and the target area is within a visible range of the pedestrian; obtaining a first recognition result by recognizing (S802, S1202) an action performed by the pedestrian, the first recognition result includes the pedestrian performs the first target action, or the pedestrian does not perform the first target action; determining (S803) a first driving policy of the vehicle based on the first recognition result; and controlling (S1203), based on that the first recognition result is the pedestrian does not perform the first target action, the target device in the vehicle to display a second request information in the target area, wherein the second request information is used to indicate the pedestrian need to perform a first road participation behavior, the first road participation behavior comprises stopping to wait, continuing to walk, walking to cross the road, or running to cross the road.
The method according to claim 1, wherein a display manner of the second request information includes time information, when the first road participation behavior is the continuing to walk, the time information is a time for the pedestrian to pass the vehicle, when the first road participation behavior is the stopping to wait, the time information is a time for the vehicle to pass the pedestrian.
The method according to claim 1 or 2, further comprises: controlling the target device in the vehicle to display a second target action that needs to be performed by the pedestrian is displayed when the second request information is displayed, the second target action is associated with the second road participation behavior of the pedestrian at a next moment.
The method according to claim 1 or 2, further comprises: obtaining a second recognition result by recognizing an action performed by the pedestrian, the second recognition result includes the pedestrian performs the first road participation behavior; determining a second driving policy of the vehicle based on the second recognition result.
The method according to claim 3, further comprises: obtaining a third recognition result by recognizing an action performed by the pedestrian, the third recognition result includes the pedestrian performs the first road participation behavior or the pedestrian performs the second target action; determining a third driving policy of the vehicle based on the third recognition result.
The method according to claim 5, further comprises: controlling, based on the first driving policy is inconsistent with the third driving policy, the vehicle to change from the first driving policy to the third driving policy.
A control apparatus, wherein the apparatus comprises a control unit, a obtaining unit, and a processing unit, wherein the control unit is configured to control (S801, S1201) a target device in a vehicle to display first request information in a target area, wherein the first request information is used to request a pedestrian to perform a first target action, and the target area is within a visible range of the pedestrian; the obtaining unit is configured to obtain a first recognition result by recognizing (S802, S1202) an action performed by the pedestrian, the first recognition result includes the pedestrian performs the first target action, or the pedestrian does not perform the first target action; and the processing unit is configured to determine (S803) a first driving policy of the vehicle based on the first recognition result, and control (S1203), based on that the first recognition result is the pedestrian does not perform the first target action, the target device in the vehicle to display a second request information in the target area, wherein the second request information is used to indicate the pedestrian need to perform a first road participation behavior , the first road participation behavior comprises stopping to wait, continuing to walk, walking to cross the road, or running to cross the road.
The apparatus according to claim 7, a display manner of the second request information includes time information, when the first road participation behavior is the continuing to walk, the time information is a time for the pedestrian to pass the vehicle, when the first road participation behavior is the stopping to wait, the time information is a time for the vehicle to pass the pedestrian.
The apparatus according to claim 7 or 8, wherein the processing unit is further configured to control the target device in the vehicle to display a second target action that needs to be performed by the pedestrian is displayed when the second request information is displayed, the second target action is associated with the second road participation behavior of the pedestrian at a next moment.
The apparatus according to claim 7 or 8, wherein the obtaining unit is further configured to obtain a second recognition result by recognizing an action performed by the pedestrian, the second recognition result includes the pedestrian performs the first road participation behavior; the processing unit is further configured determine a second driving policy of the vehicle based on the second recognition result.
The apparatus according to claim 9, wherein the obtaining unit is further configured to obtain a third recognition result by recognizing an action performed by the pedestrian, the third recognition result includes the pedestrian performs the first road participation behavior or the pedestrian performs the second target action; the processing unit is further configured determine a third driving policy of the vehicle based on the third recognition result.
The apparatus according to claim 11, wherein the processing unit is further configured to control, based on the first driving policy is inconsistent with the third driving policy, the vehicle to change from the first driving policy to the third driving policy.
A control apparatus, comprising a memory and a processor, wherein the memory stores computer program instructions, and the processor runs the computer program instructions to perform the method according to any one of claims 1 to 6.
A vehicle, comprising the apparatus according to any one of claims 7 to 12.
A computer program product, wherein when the computer program product is run on a processor, the processor is enabled to perform the method according to any one of claims 1 to 6.

Description

This application claims priority to Chinese Patent Application No. 202110574297.2, filed with the China National Intellectual Property Administration on May 25, 2021, and entitled "CONTROL METHOD AND APPARATUS", which is incorporated herein by reference in its entirety. TECHNICAL FIELD This application relates to the field of self-driving technologies, and in particular, to a control method and apparatus. BACKGROUND During vehicle driving, road participant intention determining and prediction performed by a self-driving system is a basis of route planning of the self-driving system, and is also an important condition related to road safety. Generally, when a road participant includes a pedestrian, the self-driving system may predict an intention of the pedestrian based on a movement direction of the pedestrian and the like. For example, the self-driving system may predict, based on the movement direction of the pedestrian, the intention of the pedestrian by estimating a movement track of the pedestrian according to a machine learning algorithm. However, in the foregoing prediction method based on movement of the pedestrian, a prediction result of the intention of the pedestrian may be inaccurate, and consequently driving safety is reduced. SUMMARY Embodiments of this application provide a control method and apparatus, applied to the field of self-driving technologies. The method includes: controlling a target device in a vehicle to display first request information in a target area. Because the first request information is used to request a pedestrian to perform a target action, and the target action is used to express a road participation intention of the pedestrian, a driving policy of the vehicle may be determined by recognizing an action performed by the pedestrian based on a recognition result. In this way, even without participation of a driver, the vehicle may query the road participation intention of the pedestrian by using the first request information, so that in a self-driving scenario, the vehicle can implement intention interaction with the pedestrian, to obtain an accurate road participation intention of the pedestrian, and obtain an appropriate driving policy. This improves driving safety. According to a first aspect, an embodiment of this application provides a control method. The method includes: controlling a target device in a vehicle to display first request information in a target area, where the first request information is used to request a pedestrian to perform a target action, the target action is used to express a road participation intention of the pedestrian, and the target area is within a visible range of the pedestrian; recognizing an action performed by the pedestrian; and determining a driving policy of the vehicle based on a recognition result. In this way, even without participation of a driver, the vehicle may query the road participation intention of the pedestrian by using the first request information, so that in a self-driving scenario, the vehicle can implement intention interaction with the pedestrian, to obtain an accurate road participation intention of the pedestrian, and obtain an appropriate driving policy. This improves driving safety. In a possible implementation, that the first request information is used to request a pedestrian to perform a target action includes: the first request information includes indication information used to indicate an expected action, and the expected action is associated with the road participation intention; and the determining a driving policy of the vehicle based on a recognition result includes: determining the driving policy of the vehicle based on whether the action performed by the pedestrian is the expected action. In this way, the vehicle indicates the pedestrian to perform the expected action, so that the driving policy of the vehicle can be determined based on the action performed by the pedestrian. This improves driving safety. In a possible implementation, the expected action includes a first expected action and a second expected action, the first expected action is associated with a first road participation intention of the pedestrian, and the second expected action is associated with a second road participation intention of the pedestrian; and the determining a driving policy of the vehicle based on a recognition result includes: determining the driving policy of the vehicle based on whether the action performed by the pedestrian is the first expected action or the second expected action. In this way, the driving policy of the vehicle may be determined based on whether the action performed by the pedestrian is the first expected action or the second expected action. This improves driving safety. In a possible implementation, that the first request information is used to request a pedestrian to perform a target action includes: the first request information includes indication information used to indicate a pluralit