Deception-Attack-Resilient Kinematic Control of Redundant Manipulators: A Projection Neural Network Approach

In this paper, we deal with the kinematic control problem of redundant manipulators subject to deception attacks. Specifically, the velocity command sent from the control center to the manipulator is modified by an attacker. Considering the real-time control requirement, instead of using time-consum...

Full description

Saved in:
Bibliographic Details
Published in2023 International Annual Conference on Complex Systems and Intelligent Science (CSIS-IAC) pp. 483 - 488
Main Authors Zhang, Chaofan, Zhang, Yinyan, Dai, Linyan
Format Conference Proceeding
LanguageEnglish
Published IEEE 20.10.2023
Subjects
constraint
Deception attack
End effectors
Force
kinematic control
Kinematics
Optimal control
projection neural network
Real-time systems
Recurrent neural networks
redundant manipulator
Simulation
Online AccessGet full text

Cover

More Information
Summary:In this paper, we deal with the kinematic control problem of redundant manipulators subject to deception attacks. Specifically, the velocity command sent from the control center to the manipulator is modified by an attacker. Considering the real-time control requirement, instead of using time-consuming cryptography technologies, we propose a scheme by leveraging dynamic neural networks and attack estimators. The scheme can remedy for the effect of the deception attack, deal with the joint constraints, ensure the optimality of the performance index, and force the end-effector to follow the desired trajectory. Simulation results are given to show the effectiveness of the proposed method.
DOI:10.1109/CSIS-IAC60628.2023.10364228