CN-121984732-A - Dynamic road right decision method, equipment and storage medium under network attack

CN121984732ACN 121984732 ACN121984732 ACN 121984732ACN-121984732-A

Abstract

The application discloses a dynamic road right decision method, equipment and a storage medium under network attack, which relate to the technical field of unmanned safety and comprise the steps of obtaining behavior data of a target vehicle in a preset network attack checkpoint, calculating a dynamic road right conformity score under the network attack through a preset network attack scene road right model based on the behavior data, determining a network attack resistance authority label by combining the dynamic road right conformity score and a driving license grade of the target vehicle, binding the network attack resistance authority label with an electronic driving license, comparing the network threat grade with the network attack resistance authority label, and executing corresponding road right decision by combining the dynamic road right conformity score. According to the method, by constructing a dynamic road right evaluation and arbitration mechanism integrating attack scenes, vehicle behavior performances and authority levels, the accurate depiction and real-time management and control of the decision security of Robotaxi under network attack are realized.

Inventors

WENG RENHONG
YI WEI
XIONG WEIMING
ZHANG MENG

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: 20260123

Claims (10)

1. The dynamic road right decision method under the network attack is characterized by comprising the following steps: Acquiring behavior data of a target vehicle in a preset network attack checkpoint; Calculating a dynamic road right coincidence score under network attack through a preset network attack scene road right model based on the behavior data; combining the dynamic road right conformity score and the driving license level of the target vehicle, determining a network attack resistance authority label of the target vehicle, and binding the network attack resistance authority label with a vehicle identity credential; and according to the comparison result of the received network threat level and the network attack resistance authority label contained in the vehicle identity certificate, executing corresponding road right decision by combining the dynamic road right coincidence score.
2. The method for dynamic road right decision under network attack according to claim 1, wherein the step of calculating the dynamic road right coincidence score under network attack through a preset network attack scene road right model based on the behavior data comprises: Determining a reference road right conformity score of the target vehicle under no network attack interference; calculating attack effectiveness according to the deviation degree of the key state parameters of the target vehicle before the simulated network attack injection and the vehicle after the simulated network attack injection; Determining a response compliance coefficient according to whether the actual behavior of the target vehicle accords with the network security policy response preset in the current network attack checkpoint; And calculating to obtain a dynamic road right coincidence score under the network attack through a preset multidimensional attenuation model based on the reference road right coincidence score, the attack effectiveness degree and the response compliance coefficient.
3. The method for dynamic road right decision under network attack according to claim 2, wherein before the step of calculating the dynamic road right coincidence score under network attack by using a pre-trained network attack scene road right model based on the behavior data, the method further comprises And constructing a network attack scene road weight model, wherein the network attack scene road weight model is composed of a plurality of preset network attack checkpoints.
4. The method for dynamic road right decision under network attack according to claim 1, wherein the step of determining a network attack resistant authority tag of the target vehicle by combining the dynamic road right compliance score and the driving license level of the target vehicle and binding the network attack resistant authority tag with a vehicle identity credential comprises: averaging the dynamic road right conformity scores to obtain a comprehensive network security representation value; Determining a network attack permission label of the target vehicle based on the driving license level of the target vehicle and the comprehensive network security representation value; and writing the network attack permission label into an electronic driving license of the target vehicle to generate a vehicle identity certificate containing network security qualification information.
5. The method for dynamic road right decision under network attack according to claim 1, wherein the step of executing the corresponding road right decision in combination with the dynamic road right compliance score according to the comparison result of the received network threat level and the network attack resistance authority label contained in the electronic driving license comprises: acquiring a network threat level representing the network attack risk intensity of the current road section; Based on the matching relation between the network attack resistance authority label and the network threat level and the dynamic road right coincidence degree score, executing corresponding road right decision according to a preset road right decision rule.
6. The method for dynamic road right decision under network attack according to claim 5, wherein the step of executing the corresponding road right decision based on the matching relationship between the network attack resistance authority label and the network threat level and the dynamic road right conformity score according to a preset road right decision rule comprises: comparing the network attack resistance authority label with the network threat level; If the network attack resistance authority label is greater than or equal to the network threat level, executing a corresponding road right decision based on a preset range interval where the dynamic road right coincidence degree score is located; And if the network attack resistance authority label is smaller than the network threat level, executing corresponding road right decision based on the relation between the dynamic road right coincidence degree score and a preset threshold value.
7. The method for dynamic road right decision under network attack according to claim 6, wherein if the network attack resistance authority label is greater than or equal to the network threat level, the step of executing the corresponding road right decision based on the range interval in which the dynamic road right score is located comprises: When the network attack resistance authority label is larger than or equal to the network threat level and the dynamic road right conformity score is in a first preset interval, allowing the target vehicle to normally pass; When the network attack resistance authority label is larger than or equal to the network threat level and the dynamic road right conformity score is in a second preset interval, sending a safety early warning to a target vehicle and triggering an enhanced monitoring mechanism; And triggering a network security degradation strategy when the network attack resistance authority label is larger than or equal to the network threat level and the dynamic road right conformity score is in a third preset interval.
8. The method for dynamic road right decision under network attack according to claim 6, wherein if the network attack resistance authority label is smaller than the network threat level, the step of executing the corresponding road right decision based on the relationship between the dynamic road right coincidence score and a preset threshold value comprises: When the network attack resistance authority label is smaller than the network threat level and the dynamic road right conformity score reaches the preset threshold value, issuing an authorized passing instruction to a target vehicle; And triggering a preset network security minimum risk strategy when the network attack resistance authority label is smaller than the network threat level and the dynamic road right coincidence score does not reach the preset threshold value.
9. A dynamic right-of-way decision device under a network attack, characterized in that the device comprises a memory, a processor and a computer program stored on the memory and executable on the processor, the computer program being configured to implement the steps of the dynamic right-of-way decision method under a network attack according to any of claims 1 to 8.
10. A storage medium, characterized in that the storage medium is a computer readable storage medium, on which a computer program is stored, which computer program, when being executed by a processor, implements the steps of the dynamic right-of-way decision method under network attacks according to any one of claims 1 to 8.

Description

Dynamic road right decision method, equipment and storage medium under network attack Technical Field The present application relates to the field of autopilot security technologies, and in particular, to a method, an apparatus, and a storage medium for dynamic road right decision under network attack. Background With the development of automatic driving technology, the risk of network attack is increasingly prominent, and automatic driving vehicles (especially Robotaxi) are highly dependent on complex sensors, communication systems and positioning systems to make decisions, and all the systems can be attack targets. However, the existing automatic driving test and evaluation method mainly depends on the functional safety and the behavior compliance under the conventional traffic scene, lacks a systematic and reproducible test framework aiming at the network attack scene, has the evaluation dimension more limited by whether the system is down or the function is completely disabled, ignores the robustness, decision rationality and safety degradation capability of the vehicle when the vehicle is subjected to the network attack, so that the high-safety-class vehicle with the active defense capability and the low-safety-class vehicle which can only maintain basic operation cannot be effectively distinguished, and on the other hand, the current supervision mechanism regards the network safety capability of the vehicle as a static attribute, cannot combine the response performance of the vehicle to the dynamic attack in the actual operation or standardized test to continuously evaluate and adjust the authority, causes the road right allocation mechanism to lack the perception capability of the real network safety state, and is difficult to support the refined traffic management control under the high-risk scene. Therefore, an innovative mechanism that can integrate network threat awareness, dynamic road right behavior assessment and right regulation is needed to realize accurate depiction and real-time management of Robotaxi decision security under network attack. Disclosure of Invention The application mainly aims to provide a dynamic road right decision method, equipment and a storage medium under network attack, and aims to solve the technical problem that in the prior art, the dynamic road right decision security of an automatic driving vehicle is difficult to accurately evaluate and manage in real time when the automatic driving vehicle is under network attack. In order to achieve the above objective, the present application provides a method for dynamic right-of-way decision under network attack, which includes: Acquiring behavior data of a target vehicle in a preset network attack checkpoint; Calculating a dynamic road right coincidence score under network attack through a preset network attack scene road right model based on the behavior data; determining a network attack resistance authority label of the target vehicle by combining the dynamic road right conformity score and the driving license level of the target vehicle, and binding the network attack resistance authority label with an electronic driving license of the target vehicle; and according to the comparison result of the received network threat level and the network attack resistance authority label contained in the electronic driving license, executing corresponding road right decision by combining the dynamic road right coincidence score. In an embodiment, the step of calculating the dynamic road right consistency score under the network attack through a preset network attack scene road right model based on the behavior data includes: Determining a reference road right conformity score of the target vehicle under no network attack interference; Calculating attack effectiveness according to the state deviation degree between the actual behavior of the target vehicle under the simulated network attack and the expected behavior of the target vehicle under the no network attack; Determining a response compliance coefficient according to whether the actual behavior of the target vehicle accords with the network security policy response preset in the current network attack checkpoint; And calculating to obtain a dynamic road right coincidence score under the network attack through a preset multidimensional attenuation model based on the reference road right coincidence score, the attack effectiveness degree and the response compliance coefficient. In one embodiment, before the step of calculating the dynamic road right consistency score under the network attack through the pre-trained network attack scene road right model based on the behavior data, the method further comprises the following steps of Constructing a network attack scene road weight model, wherein the network attack scene road weight model is composed of a plurality of preset network attack checkpoints, and each network attack checkpoint comprises: And the network attack vector is i