Research On Advanced Control For Networked Teleoperated Robots Under Complicated External Environments:A Switched Control Method

Since it can replace human beings for hazardous environment operations,the networked teleoperated robot has a very wide application prospect in the fields of aerospace,military,medical,disaster relief,and nuclear environments.However,the introduction of communication networks,coupled with the complicated characteristics of the robot system itself and its operating environment,has a considerable effect on the teleoperation control.During the past half century,the study of teleoperation system has received extensive attention,and many control schemes including passivity-based control and input-to-state/input-to-output stability small-gain theorem method have been investigated.Nevertheless,those methods still face many limitations when applied to the control of networked teleoperation systems.Therefore,investigating a novel effective control method is an important topic for the advanced control of teleoperation system.In this regard,this dissertation discusses several specific control design problems for teleoperation systems under complicated communication networks and external environments,which is supported by the National Natural Science Foundation of China “Fundamental Control Theory and Methods of Multi-Source Information Complex Systems”(Grant No.612250-15),and the Beijing Natural Science Foundation “Study on Minimally Invasive Spinal Surgery Telerobot System”(Grant No.4161001).By taking into account the three aspects which are the convenience of control implementation,the versatility of analysis method and the improvements on control performance,an auxiliary switched filter control method is proposed based on the concept of dynamic compensation.By utilizing the switched control theory,the closed-loop stability is established in the sense of state-independent input-to-output stability.Under a unified control framework,the dissertation has made a series of research results.Specifically,the designs and analyses for several control tasks are investigated for teleoperation systems,which include the adaptive control,the finite-time control,the saturation control,the multilateral tele-coordinated control,the semi-autonomous control and the high precision control.The main contents and results in this dissertation are summarized as follows:1.The(finite-time)adaptive control is investigated for teleoperation system in simultaneous presence of unknown asymmetric varying time delays,model uncertainties,external disturbances,and passive and/or non-passive external inputs(refer to the forces exerted on the robots by the human operator and environment).Firstly,under the proposed auxiliary switched filter framework,by introducing a filter compensation in the traditional control design,this dissertation develops a novel adaptive controller with a simple structure,i.e.,“proportional + damping injection + auxiliary switched filter compensation”.In this case,the closed-loop teleoperation system is modeled as a special class of switched nonlinear delayed system.By using the multiple LyapunovKrasovskii functionals(MLKFs)method,the closed-loop system is proved to be stateindependent input-to-output stable(SIIOS).In this scheme,the human operator/environment forces are treated as the inputs of closed-loop system,which avoids a direct discussion of their passive or non-passive characteristics.Combining with the utilization of state-independent input-to-output stability,it also overcomes the shortcomings of zero-steady-state synchronization of existing ISS/IOS small-gain methods which are often used to deal with such problem.Based on the basis,by further using the nonsmooth control technology,the finite-time adaptive control is also developed.Note that,according to the literature review,the proposed finite-time control scheme by this dissertation achieves the first theoretical proof of the finite-time control for teleoperation system with time-varying delays,which can provide faster master-slave synchronization for teleoperation systems under complicated communication networks and external environments.2.Based on the investigations above,the(finite-time)adaptive control with input saturation constraints is also developed,where two kinds of saturation control schemes are proposed.On the one hand,by intentionally introducing a nonlinear saturating function,the actual saturated controller is decoupled into two parts,i.e.,a generalized controller and the introduced saturating function.By putting the introduced saturating function on the outside of the generalized controller,the design of generalized controller will not need to consider the input saturation constraint.By making the appropriate transformation for the saturating function,and combining with the design of augmented auxiliary switched filter,the generalized controller still has a simple control structure,i.e.,“proportional + damping injection + auxiliary switched filter compensation”.By using MLKFs method,the closed-loop system is stable in the sense of state-independent input-to-output stability.On the other hand,this dissertation also investigates an anti-windup compensator based finite-time saturated adaptive controller.In this case,firstly,to handle the actuator saturation,a novel anti-wind compensator is developed.Then,it is applied to design the auxiliary switching filter as well as the adaptive controller.By further using the non-smooth control technology,the closedloop system is shown to be finite-time SIIOS(FTSIIOS).3.By utilizing the auxiliary switched filter control method,this dissertation also addresses the coordinated control of multi-master-multi-slave teleoperation systems,where two modified tele-coordinated control schemes are investigated.On the one hand,a hybrid motion/force control based tele-coordinated control is developed for a multilateral asymmetric teleoperation with multiple slave mobile manipulators.In this issue,firstly,by using space transformation theory,the decoupled space of motion and force is established,which obtains the concise motion model and force model.Then based on the auxiliary switched filter control method,an adaptive fuzzy controller is developed,which guarantees the tele-coordinated control of multiple mobile manipulators in a less conservative manner.On the other hand,to improve the control performance of conventional non-constrained coordinated control schemes,this dissertation also investigates a novel motion synchronization of slave robots based non-constrained tele-coordinated controller.By introducing the additional prescribed performance synchronization control of slave robots in the conventional non-constrained coordinated controller,a novel auxiliary switched filter based adaptive neural network coordinated controller is developed,which improves the coordination performance between the slave robots in the presences of the different human operators' actions and external uncertainties(refer to different communication delays and model uncertainties).In comparison with the hybrid motion/force control based tele-coordinated control,this scheme reduces the use of force sensors while ensuring good coordination performance.And thus,it reduces the economic and technical burden of the application.4.In this dissertation,the auxiliary switched filter control method is also adopted to the high flexibility and high precision control of minimally invasive spinal surgery telerobot system.(a)To achieve the high flexibility of minimally invasive spinal surgery telerobot system,this dissertation addresses the semi-autonomous control of heterogeneous teleoperation system with redundant slave robot.By using auxiliary switched filter control method,a novel semi-autonomous adaptive controller is developed.The proposed control not only ensures the motion synchronization between the master and slave robots,but also achieves the additional sub-tasks control including avoiding singularity,joint limits,and avoiding obstacle for the redundant slave robot,in the presence of time-varying delays,passive/non-passive exogenous forces,dynamics parameter uncertainties and input uncertainties.And thus,it improves the operating capacity of teleoperation system in complicated and dynamic environment.(b)For the high precision control requirement,this dissertation develops two high performance control methods by using the classical prescribed performance control(PPC)and its modified strategy.Firstly,by combining the auxiliary switched filter control method with the classical PPC scheme,the quantitative analysis and control of synchronization performance(includes the synchronization speed,the maximum overshoot and steady-state error)between the master and slave robots with time-varying communication delays are obtained,no matter whether the robots contact with the human operator/environment or not.After that,a switching-based prescribed performance-like control(PPLC)approach is also investigated for nonlinear teleoperation systems.Attention is focused on how to guarantee the control implementability and the control performance when the hard constraint of classical PPC algorithm is violated,since the violation of the hard constraint of a classical PPC scheme can lead to the control singularity and then cause closed-loop instability.It is proved that the developed switching-based PPLC approach improves the environmental adaptability of the control algorithm,with retaining the similar excellent performance of the classical predefined performance control algorithms.Note that,for different control tasks above,all of the designed controllers based on the auxiliary switched filter control method have the general form of “proportional + damping injection + auxiliary filter compensation”.The effectiveness of all of the proposed control algorithms is demonstrated by the experimental results or numerical simulations.Compared with the existing control methods,the proposed algorithms of this dissertation have achieved a better control performance in a wider range of applications,where the controller structure is more concise,the stability analysis is easier to transplant,and the conservativeness is lower.At the end of this dissertation,the main results are concluded and the perspectives of future researches are presented.