15743

1618 دكتري

دكتري

كنترل

اصفهان : دانشگاه صنعتي اصفهان

1399

سيزده، 122ص.: مصور، جدول، نمودار

ايمان ايزدي، فريد شيخ الاسلام

جعفر قيصري، محمد دانش، حميد خالوزاده

1399/06/19

كتابنامه

برق

مهندسي برق و كامپيوتر

1399/06/24

2630164

Design of Robust Guidance Law for Multi Agent Systems with Input Saturation and Disturbance Saeed Khankalantary Roknabadi s khan@ec iut ac ir Auguest 2020 Department of Electrical and Computer Engineering Isfahan University of Technology Isfahan 84156 83111 Iran Degree Ph D Language FarsiSupervisor Assoc Prof Iman Izadi iman izadi@cc iut ac ir Supervisor Prof Farid Sheikholeslam sheikh@cc iut ac ir Abstract In this research a new framework is studied and presented to design a guidance law for a multi agent interceptionsystem which is used to tackle high speed targets These targets are not easily recognizable so accurate estimation of theirposition is not available at far distances and measurement systems only detect a range of target presence On the otherhand due to various uncertainties in the dynamical modeling of the interceptors the designed guidance and control lawsmust be To deal with the uncertainty of target identification according to the statistical properties of the measurementsystems we consider the target as an envelope To tackle this envelope we define a virtual hull on the interceptor side which is used to guide the agents in the middle phase In this phase the points that are optimally selected are placed inthe virtual hull and each agent is directed to these points To do this first the problem of extracting these optimal pointsis investigated and by solving it the optimal formation is extracted inside the hull Then the multi agent guidance law isdesigned with the differential game approach using dynamic adaptive programming The problem is constrained by collisionavoidance input saturation and disturbance Integral reinforcement learning is used for training In this approach eachagent follows a path that results in multi agent formation tracking Next due to the importance of increasing efficiency optimal design and cost effectiveness an integrated guidance and control law is designed In this method considering thedisturbance constraints parametric uncertainty and input saturation an integrated guidance and control law is proposedusing extended state observer and command filter Unlike the mid phase the final phase is functionally single agent Inthe end an integrated simulation for high speed target tracking is performed in which the mid phase and final phase areimplemented using the proposed integrated guidance and control law It is shown that the proposed framework improvestarget interception Key Words Robust Guidance Multi Agent System Adaptive Dynamic Programming Disturbanceand Input Saturation Integrated Guidance and Control

ايمان ايزدي، فريد شيخ الاسلام

جعفر قيصري، محمد دانش، حميد خالوزاده