CN-121536292-B - Intention recognition method, apparatus, device, medium, and program product

CN121536292BCN 121536292 BCN121536292 BCN 121536292BCN-121536292-B

Abstract

The application provides an intention recognition method, device, equipment, medium and program product, which are used for determining a target curb based on a lane where a target vehicle is located, determining a road participant which is positioned at the outer edge of a target road and is used as a target road participant based on the position of the target curb, and predicting the behavior intention of the target road participant based on the road participant category corresponding to the target road participant and the curb category corresponding to the target curb to obtain the target behavior intention representing whether the target road participant spans the target curb. Therefore, based on the curb category which can represent the straddlability of the curb and combining the road participant categories corresponding to different movement modes, the accurate prediction of the behavior intention of the target road participant can be realized under the condition of comprehensively considering the actual scene, and the judgment error is avoided, so that the safe passing of vehicles is controlled based on the behavior intention, the collision risk is effectively reduced, and the passing efficiency is improved.

Inventors

ZHENG NIANNIAN
LUO YUBAO
WEI CHAO

Assignees

浙江吉利控股集团有限公司
吉利汽车研究院（宁波）有限公司

Dates

Publication Date: 20260505
Application Date: 20260120

Claims (10)

1. A method of intent recognition, the method comprising: Determining a target curb based on a lane in which a target vehicle is located; determining a target road participant based on a location of the target curb, the target road participant comprising a road participant located outside the target curb; predicting the behavior intention of the target road participant based on the road participant category corresponding to the target road participant and the curb category corresponding to the target curb to obtain the target behavior intention of the target road participant, wherein the road participant category corresponding to the target road participant represents the movement characteristics of the target road participant, the curb category of the target curb represents the crossing property of the target curb, and the target behavior intention represents whether the target road participant crosses the target curb.
2. The intention recognition method according to claim 1, wherein before predicting the behavior intention of the target road participant based on a road participant class corresponding to the target road participant and a curb class corresponding to the target curb, the method further comprises: And determining a curb category corresponding to the target curb based on whether the target curb is a physical entity and a preset curb height.
3. The method for identifying an intention according to claim 2, wherein the determining a curb category corresponding to the target curb based on whether the target curb is a physical entity and a preset curb height comprises: if the target curb is not a physical entity, determining that the curb type corresponding to the target curb is a virtual curb; if the target curb is a physical entity, determining that the curb type corresponding to the target curb is a high-physical curb or a low-physical curb based on the magnitude relation between the preset curb height and the target curb height.
4. The intention recognition method according to claim 3, wherein predicting the behavior intention of the target road participant based on the road participant class corresponding to the target road participant and the curb class corresponding to the target curb to obtain the target behavior intention of the target road participant comprises: If the curb class corresponding to the target curb is a virtual curb, predicting the behavior intention of the target road participant based on a target image to obtain the target behavior intention, wherein the target image comprises the target road participant; If the curb class corresponding to the target curb is a low-physical curb, predicting the behavior intention of the target road participant based on the type of the road participant corresponding to the target image and the target road participant to obtain the target behavior intention; If the curb class corresponding to the target curb is a high-physics curb, determining that the target behavior is intended to characterize that the target road participant does not cross the target curb.
5. The intention recognition method according to claim 4, wherein predicting the behavior intention of the target road participant based on the target image to obtain the target behavior intention includes: identifying a target orientation of the target road participant in the target image; And determining the target behavior intention based on the magnitude relation between a target included angle and a preset included angle, wherein the target included angle is an included angle between the target orientation and the target curb.
6. The method for recognizing intention according to claim 4, wherein predicting the intention of the behavior of the target road participant based on the road participant type corresponding to the target road participant in the target image, to obtain the target intention of the behavior, comprises: If the road participant type corresponding to the target road participant is a pedestrian, determining that the target behavior intention represents that the target road participant can cross the target curb; If the road participant type corresponding to the target road participant is a motor vehicle, determining that the target behavior intention represents that the target road participant cannot cross the target curb; and if the road participant type corresponding to the target road participant is a non-motor vehicle, predicting the behavior intention of the target road participant based on the target image to obtain the target behavior intention.
7. An intent recognition device, the device comprising: the first determining module is used for determining a target curb based on a lane where a target vehicle is located; A second determining module for determining a target road participant based on a location of the target curb, the target road participant comprising a road participant located outside the target curb; The prediction module is used for predicting the behavior intention of the target road participant based on the road participant category corresponding to the target road participant and the curb category corresponding to the target curb to obtain the target behavior intention of the target road participant, the road participant category corresponding to the target road participant represents the movement characteristic of the target road participant, the curb category of the target curb represents the straddlability of the target curb, and the target behavior intention represents whether the target road participant spans the target curb.
8. An electronic device comprising a memory and a processor; the memory is connected with the processor and used for storing programs; The processor is configured to implement the intention recognition method according to any one of claims 1 to 6 by running a program in the memory.
9. A storage medium having stored thereon a computer program which, when executed by a processor, implements the method of intent recognition as claimed in any one of claims 1 to 6.
10. A computer program product comprising computer program instructions which, when executed by a processor, cause the processor to implement the method of intent recognition as claimed in any one of the preceding claims 1 to 6.

Description

Intention recognition method, apparatus, device, medium, and program product Technical Field The application relates to the technical field of vehicle driving, in particular to an intention recognition method, an intention recognition device, an intention recognition medium and a program product. Background Along with the trend of automatic driving technology from theoretical research to large-scale chemical engineering application, two major functions of memory parking (Home-zone PARKING ASSIST, HPA) and city navigation assisted driving (CITY NAVIGATE on Autopilot, CNOA) become two major keys for determining product competitiveness, and the two major functions support automatic driving experience of vehicles in complex scenes such as cities, parks and the like. In CNOA applications, vehicles are subjected to a variety of scenarios, especially in some edge scenarios (Corner Case), i.e. more complex driving scenarios, such as intra-campus roads, crossroads without lane markers, areas where cell entrances and exits engage city thoroughfares, rural roads, zebra crossing circumferences, etc., where the behavior intent of the dynamic proxy outside the curb near the curb is difficult to determine, wherein the proxy is based on data models or software objects created by road participants, which are pedestrians, non-motor vehicles (e.g. two-wheelers), motor vehicles, etc. Because the dynamic proxy can cross the curb or invade the current driving lane of the vehicle, if the behavior intention of the dynamic proxy is mispredicted, the collision risk is caused by the fact that the vehicle is not avoided, or the traffic efficiency loss is caused by the fact that the vehicle is excessively avoided. Therefore, how to accurately determine the behavior intention of the dynamic agent on the outer side of the curb, which is close to the curb, is a problem to be solved. Disclosure of Invention Based on the defects and shortcomings of the prior art, the application provides an intention recognition method, an intention recognition device, an intention recognition medium and an intention recognition program product, which can accurately predict whether a vehicle has a behavior intention of crossing a curb, so that collision risk is reduced, and vehicle passing efficiency is ensured. According to a first aspect of the present application, there is provided an intention recognition method comprising: Determining a target curb based on a lane in which a target vehicle is located; Determining the target road participant based on the location of the target curb, the target road participant comprising a road participant located outside the target curb; Predicting the behavior intention of the target road participant based on the road participant category corresponding to the target road participant and the curb category corresponding to the target curb to obtain the target behavior intention of the target road participant, wherein the target behavior intention characterizes whether the target road participant can cross the target curb. According to a second aspect of an embodiment of the present application, there is provided an intention recognition apparatus including: the first determining module is used for determining a target curb based on a lane where a target vehicle is located; A second determination module for determining the target road participant based on a location of the target curb, the target road participant comprising a road participant located outside the target curb; The prediction module is used for predicting the behavior intention of the target road participant based on the road participant category corresponding to the target road participant and the curb category corresponding to the target curb to obtain the target behavior intention of the target road participant, and the target behavior intention characterizes whether the target road participant can cross the target curb. According to a third aspect of embodiments of the present application, there is provided an electronic device comprising a memory and a processor; the memory is connected with the processor and used for storing programs; the processor is configured to implement the intention recognition method according to the first aspect by running a program in the memory. According to a fourth aspect of embodiments of the present application, there is provided a storage medium having stored thereon a computer program which, when executed by a processor, implements the method of intent recognition as described in the first aspect. According to a fifth aspect of embodiments of the present application, there is provided a computer program product, which when executed by a processor, causes the processor to implement the method of intent recognition as described in the first aspect. In the intention recognition method, the device, the equipment, the medium and the program product, the intention of the target road participant can be predicted by determinin