|
|
Actuator Attack Detection for Autonomous Vehicles |
HAN Xiao, ZHANG Mengzhen, WU Yi, CUI Xiaokai, QIU Changbin, WANG Qingzhi, LIU Qipeng
|
Institute of Complexity Science, Qingdao University, Qingdao 266071, China |
|
|
Abstract This paper proposes a method to detect the compromised actuators and then determine the attack types. Based on sensor measurements, the actual control command conducted by the actuator can be estimated using maximum likelihood estimation method; and then by comparing the estimated command with the reference one which is sent from the on-board computer, we are able to figure out whether the control command is successfully conducted by the actuator or not. Experimental results based on simulation platform show that the detector can effectively detect error once the attack occurs and precisely identify the specific type of the sensor attack.
|
Received: 02 May 2021
Published: 12 October 2022
|
|
|
|
|
|
|
|