SVM-Based Fault Type Classification Method for Navigation of Formation Control Systems
In this paper, we propose a fault type classification algorithm for a networked multi-robot formation control. Both actuator and sensor faults of a robot are considered as node fault on the networked system. The Support Vector Machine (SVM) based classification scheme is proposed in order to classify the fault type accurately. Basically, the graph-theoretic approach is used for modeling the multi-agent communication and to generate the formation control law. A numerical simulation is presented to confirm the performance of proposed fault type classification method.
KeywordsFault type classification Networked multi-robot/agent Formation control Support vector machine (SVM) Graph Theory
This work was supported by the ICT R&D program of MSIP/IITP. [R-20150223-000167, Development of High Reliable Communications and Security SW for Various Unmanned Vehicles].
- 5.Kwon, C., Liu, W., Hwang, I.: Security analysis for cyber-physical systems against stealthy deception attacks. American Control Conference (ACC), 2013. IEEE (2013)Google Scholar
- 6.Negash, L., Kim, S.-H., Choi, H.-L.: Distributed unknown-input-observers for cyber attack detection and isolation in formation flying UAVs (2017), arXiv:1701.06325
- 12.Isermann, R.: Model-based fault-detection and diagnosisstatus and applications. Annu. Rev. control 29(1), 71–85 (2005)Google Scholar
- 14.Beck, J.V., Arnold, K.J.: Parameter Estimation in Engineering and Science. James Beck (1977)Google Scholar