KR20170086302A - Method for stablizing multilateral teleoperation and system therefor - Google Patents

Abstract

10. A system for stabilizing remote control between a plurality of terminals, the plurality of terminals comprising a plurality of first type terminals and at least one second type terminal, the system comprising: Type terminal, the network receiving input energy from the first type terminal and transmitting the input energy to the second type terminal, receiving input energy from the second type terminal, and transmitting the input energy to the first type terminal , network; A passivity observer for monitoring the passivity of the network; And a passivity controller for releasing energy such that the network maintains passivity, wherein the passivity observer compares the input energy generated by all of the terminals except the terminal i among the plurality of terminals, i, respectively, and the passivity controller adjusts the amount of energy output to the terminal i based on the monitoring result.

Description

[0001] METHOD FOR STABLIZING MULTILATERAL TELEOPERATION AND SYSTEM THEREFOR [0002]

The present disclosure relates to a method for stabilizing multipoint remote control and a system therefor. More particularly, this disclosure relates to a method and system for ensuring the passivity of a network connecting multiple terminals to stabilize multipoint remote control.

Over the past few decades, remote control technology has steadily increased in applications such as space exploration, underwater business, mining, nuclear and poisoning, surgery, and entertainment businesses. Remote control technology combines human intelligence with repeatability and robustness of the power of a robot, allowing manipulation skills such as human beings to be applied to environments that are too dangerous and difficult for people to access. These advantages can be realized through coordinated control of the master robot and the slave robot. The movement of the master robot commanded by the human operator is transmitted to the slave robot, and the slave robot is controlled to follow the master operation command as it is. The interaction force of the slave robot, which is calculated when the slave robot operates the surrounding environment, is transmitted to the master robot, and the master robot is controlled to simulate such feedback force.

Recent researchers have shown great interest in the advantages of a single remote control compared to a single remote control, which provides improved maneuverability and higher loading capacity and better collaboration among people than a single remote control. . In the group remote control, operators can control multiple slave robots through multiple master robots using various command strategies. For example, each master may exclusively command each of the slaves, or several masters may command the same slave together, or a single master may command multiple slaves. Therefore, the limitations of traditional remote control applications as well as emerging areas such as haptic control, virtual shopping, virtual prototype haptic information sharing, and surgical training through grouped remote control, It is expected to be able to solve it.

I will add it after I confirm the claim.

1 shows a conceptual diagram of a system for remote control with a plurality of masters and a plurality of slaves.
Figure 2 shows a block diagram of a bilateral remote control system of a position-force bilateral control architecture with (measured) communication time-delay.
Figure 3 shows the TDPN representation of Figure 2;
Figure 4 shows the energy flows in the TDPN.
5A is a circuit diagram illustrating a time delayed power network handled using a passivity observer and a passivity controller in an impedance configuration.
5B is a circuit diagram illustrating a time delayed power network handled using a passivity observer and passivity controller in an admittance configuration.
Figure 6 shows all possible signals of a multipoint remote control system.
Figure 7 shows a separate, multi-way remote control system.
8 is a network representation of an i < th > terminal with multiple power and velocity sources.
Figure 9 is a network representation of an i < th > terminal that considers only a plurality of receive sources.
10 is an enhanced network representation of an i < th > terminal with multiple receive sources.
FIG. 11 shows a network representation in which only the velocity sources are considered in FIG.
12 is an augmented network representation of an i-th terminal with multiple rate sources.
Figure 13 is a final network representation of a multipattern remote control system decoupled for a terminal.
Figure 14 shows the resultant animated architecture of the i-th terminal of a multipoint remote control system via TDPA.
Figure 15 shows the i-th terminal that is less conservatively manipulated using TDPA.
Figure 16 shows the energy flows of a multipoint remote control system.
Figure 17 shows the decoupled energy relationship in a multipoint remote control system.
18 shows a three-way remote control system.
19 shows three network representations decoupled of the three-way telemetry system in FIG. 18 with TDPNs and POs / PCs.
Figure 20 shows a graph showing the unstable position / force response without the proposed multipurpose controller.
Figure 21 shows a comparison of input and output energies at each TDPN without the proposed multipoint controller.
Figure 22 shows a graph showing a stable position / force response using the proposed multipurpose controller.
23 shows a comparison of the input and output energies at each TDPN using the proposed multipoint controller.
Figure 24 shows a less conservative version of the passive technique for Master 1 and Slave.
Figure 25 shows a graph showing a stable position / force contact response using the proposed less conservative multi-zone controller.
26 shows a comparison of input and output energies at the output terminal using the proposed less conservative multi-party controller.

Hereinafter, an apparatus and method for improving performance in remote control based on time domain passiveness according to the present invention will be described in detail with reference to the accompanying drawings. The present invention is not limited to the above-described embodiments, and various changes and modifications may be made without departing from the scope of the present invention. In addition, the matters described in the attached drawings may be different from those actually realized by schematically illustrating the embodiments of the present invention.

In the meantime, each constituent unit described below is only an example for implementing the present invention. Thus, in other implementations of the present invention, other components may be used without departing from the spirit and scope of the present invention.

In addition, in the entire specification, when a part is referred to as being 'connected' to another part, it is not always the case that the part is directly connected to the other part, . The expression " comprising " is intended to be merely an indication of the presence of that element as an open-ended representation, and should not be understood as excluding the additional elements.

Also, the expressions such as 'first, second', etc. are used only to distinguish between plural configurations, and do not limit the order or other features among the configurations.

In addition, although these teachings are described in terms of various embodiments, these teachings are not intended to be limited to such embodiments. Rather, as will be appreciated by one of ordinary skill in the art, these teachings include various alternatives, modifications, and equivalents.

I. Overview

1 shows a conceptual diagram of a system for remote control with a plurality of masters and a plurality of slaves. Such a system may be referred to as a multiparty remote control system since position and force signals are exchanged between three or more terminals that may be master or slave.

So far, despite the high possibility of group remote control, there have been few discussions about the multi-party remote control. Sirouspour used μ-synthesis to design a multiparty control architecture with multiple slaves holding common tools for manipulating common environments, and using adaptive nonlinear controllers to accommodate nonlinearity and parametric uncertainties in the system I extended my ideas further. Khademian and H-Zaad introduced a 4-channel and 6-channel multipoint shared controller for dual-user remote control systems by introducing a dominance factor that allows adjustable interaction with each user and other users as well as slaves Respectively. However, the time-delay problem was not considered at all in the above-mentioned studies. Kanno and Yokokohji used a wave-variable method to overcome the time-delay problem in a multipoint remote control system, and by adjusting the priority of each operator by mitigating force balance constraints, . However, these methods were limited to a position-to-position bilateral control architecture.

Overall, there were some controlled approaches to attempting to stabilize the time-delayed multilateral remote control system. However, these controllers were designed for a given multipoint control architecture and were not applicable for a general multiport control architecture, because the controller had to be completely redesigned when the multiparty control architecture changed or the number of master / slaves changed. Stable multi-party control for any time-delay, any multiparty control architecture, and any master / slave count has not yet been established.

On the other hand, there has been much research on the bilateral remote control based on the Time Domain Passivity Approach (TDPA). TDPA-based remote control has several advantages over other control methods, such as ensuring passivity for an arbitrary length of time-delay, requiring no knowledge of the dynamics of the connected master and slave, Lt; / RTI > architecture. It is very difficult to design the controller by considering all possible combinations of masters and slaves in advance, which is essential for the design of a multipurpose remote control system.

The main contribution of this disclosure is to propose a general TDPA-based framework that ensures reliable remote control independent of the amount of time-delay, the multipoint control architecture and the number of master / slaves.

A major aspect of the proposed framework is how the conventional representation of a multiparty remote control system can be changed to a network representation for the implementation of the well-developed TDPA framework. A network representation with clear energy flows is essential for the implementation of TDPA. However, it has not been easy to change a multipoint remote control system from a conventional block diagram domain to an electrical circuit of a network domain having energy flows. The main challenge was the combined nature of the multilateral remote control system. This disclosure describes a general framework for network translation of a multiparty remote control system and how to secure the passivity of the network system converted to TDPA. It also provides a more comprehensive theorem with proof of the passivity of the whole system. In addition, less conservative passivity techniques will be introduced, and performance comparison studies will also be conducted.

II. Review of TDPA-based bilateral remote control

This section briefly describes the TDPA-based bilateral teleprocessing framework of position-to-force bilateral control architecture (measured). Figure 2 shows a block diagram of a bilateral remote control system of a position-force bilateral control architecture with (measured) communication time-delay. The master transmits the speed command to the slave, which feeds back the interaction force measured through the communication network to the master.

The most important thing in designing a TDPA-based bilateral controller is to identify a power conjugate pair in the communication channel. The most common mistakes associated with this architecture are f _e (measured at the slave) and v _sd at the right (slave side) of the network (The target speed command to the slave) into a power conjugate pair. This pairing appears to be correct because it appears to be the result of v _{sd on} the block diagram due to f _e , but in terms of energy calculation, this pairing is not appropriate. First, these signals represent non-juxtaposed pairs, and secondly, the measured force may remain at zero even if v _sd has a non-zero value, which is not clear from an energy standpoint.

The above-described uncertain causality can be solved using a mechanical-electrical analogy technique. Based on this analogy technique, the master and slave, the slave controller, the human operator and the remote environment can be described as passive electrical circuit elements, and the velocity signal and the force signal are respectively the flow signal, the effort signal . ≪ / RTI > However, passive electrical components can not represent communication channels with time-variable delays, and time-varying delays can potentially introduce activeness into the network. To solve this problem, the TDPA framework has been proposed to maintain the passivity of delayed communication channels. The TDPA framework has been extended by proposing a Time Delay Power Network (TDPN) framework to solve the unclear network causality when the measured power or location through the delayed network channel is transmitted as a feedback signal from the slave.

Figure 3 shows the TDPN representation of Figure 2; The idea behind the TDPN formulation is to identify the underlining sources of the command and feedback signals and express them as the corresponding ideal flow or effort source. In Fig. 3, the delayed communication channel is represented by two TDPNs, one connected to the master side and the other connected to the slave side. Based on the definition of passivity, if all the components of the network are passive, the whole system becomes passive. In the transformed network representation (FIG. 3), all components except two TDPNs are considered to be passive. Therefore, ensuring the passivity of the two TDPNs is a sufficient condition for securing a stable bilateral remote control. TDPA can be used to move the delayed network.

TDPA consists of two main components: Passivity Observer (PO), which monitors the energy flow of the network in real time, and Passivity Controller (PC), which emits the amount of active energy flowing in the network.

)의 흐름은 다음과 같이 나타낼 수 있다. Figure 4 shows the energy flows in the TDPN. Referring to FIG. 4, the time delayed power network energy (

) Can be expressed as follows.

는 제 1 포트(예를 들어, 마스터측 포트)에서의 에너지이고,

는 제 2 포트(예를 들어, 슬레이브측 포트)에서의 에너지이며, 이들은 다음과 같이 나타낼 수 있다.here,

Is the energy at the first port (e.g., the master side port)

Is the energy at the second port (e.g., the slave-side port), which can be expressed as follows.

는 샘플링 기간이다. here,

Is a sampling period.

The energy at the first port and the second port can be expressed by the input / output energy as follows.

That is, the passivity condition of the time delay power network can be expressed as follows.

및

는 통신 시간-지연 때문에 동시에 관찰될 수 없기 때문에, 상기 수동성 조건은 실시간으로 체크될 수 없다. 따라서, 이를 해결하기 위해, 상기 수학식 3 및 상기 수학식 4를 결합하면, 수동성 조건을 다음과 같이 나타낼 수 있다. At this time,

And

The passivity condition can not be checked in real time because it can not be observed simultaneously due to communication time-delay. Therefore, in order to solve this problem, the combination of Equation (3) and Equation (4) can express the passivity condition as follows.

과

은 좌측 포트(제 1 포트)에서 우측 포트(제 2 포트)로 그리고 우측 포트(제 2 포트)에서 좌측 포트(제 1 포트)로의 디커플링된 에너지 흐름들이다. 제 1 포트(좌측 포트)에서 제 2 포트(우측 포트)로 전달된 에너지 및 제 2 포트(우측 포트)에서 제 1 포트(좌측 포트)로 전달된 에너지는 다음과 같이 나타낼 수 있다. here,

and

Are the decoupled energy flows from the left port (first port) to the right port (second port) and from the right port (second port) to the left port (first port). The energy transferred from the first port (left port) to the second port (right port) and the energy transferred from the second port (right port) to the first port (left port) can be expressed as follows.

,

이면 만족할 수 있다. 따라서, 시간 지연을 고려하여, 관찰가능한 수동성 조건은 다음과 같이 나타낼 수 있다. Here, the passivity condition is

,

It can be satisfied. Therefore, considering the time delay, the observable passivity condition can be expressed as follows.

및

는 제 1 포트에서 제 2 포트로 그리고 제 2 포트에서 제 1 포트로의 통신 지연의 양을 나타낸다. 즉, 수동성 조건은 물리적으로, 한 포트에서의 출력 에너지가 반대편 포트에서의 입력 에너지를 상한으로 하여 항상 제한받는 것을 의미할 수 있다.here,

And

Represents the amount of communication delay from the first port to the second port and from the second port to the first port. That is, the passivity condition may physically mean that the output energy at one port is always limited to the upper limit of the input energy at the opposite port.

Hereinafter, the time delayed power network that is operated using the passivity observer and the passivity controller will be described in detail with reference to FIG.

A. Passivity Observer (PO)

,

)을 다음과 같이 나타낼 수 있다. The passivity observer is a component that computes Equation 7, which is a passivity condition in real time. The energy monitored by the two passivity observers for each port in the time delay power network

,

) Can be expressed as follows.

및

는 수동성 제어기에 의해 교정되는 에너지 값들이다. here,

And

Are the energy values calibrated by the passivity controller.

B. Passivity controller (PC)

The passivity controller is an adaptive damping element that limits the output energy at one port to the input energy from the opposite port by emitting an active energy when any excess energy, active energy, is monitored at the output port by the PO .

Referring to FIG. 5A, in the impedance configuration, the passivity controller modifies the force at a speed as follows.

는 슬레이브로부터 송신되는 지연된 측정 힘 신호로,

이다. 또한, 상기

는 가상 댐핑 값으로서 다음과 같이 계산될 수 있으며,Here,

Is a delayed measurement force signal transmitted from the slave,

to be. In addition,

Can be calculated as a virtual damping value as follows,

The energy release by the passivity controller can be expressed as:

는 시간 지연 전력 네트워크로부터 나오는 힘 신호를 수정하기 위해 사용될 수 있다. On the other hand, in the electric circuit, the passivity controller can be expressed as a variable resistor connected in series or in parallel depending on the impedance or admittance configuration. However, in the drawings of the present specification, the passivity controller is represented by a block diagram for convenience. For reference, in the impedance configuration,

May be used to modify the force signal coming from the time delayed power network.

Referring to Fig. 5B, in the admittance configuration, the passivity controller can be expressed as follows.

는 마스터로부터의 지연된 속도 신호로서, 이다. 또한, 상기

는 가상 댐핑 값으로서, 다음과 같이 계산될 수 있으며,Here,

Is a delayed speed signal from the master, to be. In addition,

As a virtual damping value, can be calculated as follows,

The energy emitted by the passivity controller can be expressed as:

는 시간 지연 전력 네트워크로부터 나오는 속도 신호를 수정하기 위해 사용될 수 있다. 한편, 수동성 제어기는 이상적인 종속 소스들에 연결된 포트에는 필요하지 않음을 주의한다.For reference, in the admittance configuration,

May be used to modify the rate signal from the time delayed power network. Note, however, that passivity controllers are not required for ports connected to ideal dependent sources.

III. Network Representation of Multilateral Remote Control System

As described above, the most basic step in the design process of the TDPA is to convert the block diagram of a conventional target system into a network representation. In this section, we propose a method of converting a generic block diagram of a multiparty remote control system into a network representation, which will enable us to implement a well-developed TDPA framework. It is necessary to track the flow of energy from terminal to terminal so that the terminal identifies the contribution of energy to other terminals contributing to network representation. However, in a multipurpose remote control system, it is not easy to grasp the energy contribution clearly due to the coupled nature of the multipurpose remote control system. A single input energy from one terminal may affect multiple output energies of multiple terminals at the same time or multiple input energies from multiple terminals may generate an enhanced output energy for a single terminal . For example, the measured interaction force from a single slave may be sent to multiple masters as feedback forces, or an augmented speed input from multiple masters may generate a control force output to a single slave . The input and output energies can be estimated from the velocity and force signals after the power conjugate pairs of velocity and force signals, respectively, are identified.

Figure 6 shows all possible signals of a multipoint remote control system. At each terminal, velocity and force signals can enter the network and exit the network. v _in ^M1 denotes the input speed signal from the master 1 to the network, and _vout ^S1 denotes the output speed signal from the network to the slave 1. There are a variety of possible combinations in signal exchange, and these depend on the control architecture. For example, the combination of input speeds from master 1 and master 2 may be the speed output to slave 1, and the combination of input forces from slave 1 and slave 2 may be an output force to master 2. Thus, the output energy at the terminals is generally the result of the coupled input signals from the other side of the network. The main difficulty in identifying energy contributions is the coupled nature of the multilateral remote control system.

In order to identify the energy contributions, it is necessary to identify the contribution of the input signals to each terminal. It can be assumed that it is always possible to express the output speed and force at the terminal as the sum of the contributions of the input signals from the other side of the network and the velocity output of the slave 1 and the force output of the master 2 may be as follows.

Where v _out ^S1 is the output speed from the network to slave 1, and f _out ^M2 is the output power from the network to master 2. v _in ^M1toS1 silver Is the velocity contribution of master 1 to slave 1, and v _in ^S2toS1 is the velocity contribution of slave 2 to slave 1. Other expressions follow the same notation rules. Because it depends on a given multiparty control architecture, it is always possible to identify these contributions without losing generality.

The reason hidden behind the representation of the output signals at the terminal by the contribution of the input signals from the other side of the network lies in te coupling of a multipurpose remote control system to the other terminal, . By using Equation (15) as an example, it is possible to decouple the multipoint remote control system for each output terminal.

FIG. 7 illustrates separate, multiple remote control systems. When mechanical-electrical analogy is used to map the force to the voltage with the speed as the current, the decoupled subsystems as in FIG. 7 can be converted into the electrical circuit as in FIG. 8 is a network representation of an i < th > terminal with multiple power and velocity sources. Electrical circuit transformation shows clear power conjugate pairs and energy flows. The output speed and force signals to the terminal are mapped to the target speed v _di and the feedback force f _di , which can be expressed as a port source in slave current and slave. V _i is the resulting velocity at terminal i, Z i is the impedance of the device at terminal i, K _p and K _d are proportional and inductive position controller gain, f _ci is the resulting position controller force, f _ei is i Lt; / RTI > terminal. From now on, the ith notation will be used without distinction between master and slave to reduce the length of the mathematical expression.

The target rate for terminal i is the sum of the contributions of the velocities from the other side of the network, and each contribution is generally regarded as a function of the respective velocities as follows.

는 통신 네트워크 후에 v_p의 지연된 신호이다. 유사하게, 피드백 힘은 네트워크의 다른 측으로부터의 힘의 함수들의 합으로 다음과 같다.Where h _p (v _p ) is a function of the velocity signal from terminal p,

Is a delayed signal of v _p after the communication network. Similarly, the feedback force is the sum of the functions of the force from the other side of the network as follows.

는 터미널 q로부터의 지연된 힘 신호

의 함수이다.here,

Lt; RTI ID = 0.0 > q < / RTI &

.

So far, TDPA has only considered one-to-one energy mapping, from one terminal to another, but one-to-one energy mapping is not suitable for representing a multipurpose remote control system. Thus, in this disclosure, many-to-one energy mapping is first introduced in TDPA.

For an output terminal, each subsystem is represented by a network circuit, and due to the representation in the network circuit, at least the energy contribution at the output side of the terminal becomes clear. However, multiple signal sources having different amounts of time-delay can still not be clearly expressed. As shown in FIG. 8, there are two different types of augmented sources including all the different time-delays values. Thus, the system is likely to be active and unstable due to active energy generated from delayed dependent sources. On the other hand, TDPA can emit active energy and make the system passive and stable. However, since the network representation of FIG. 8 does not explicitly represent delayed communication blocks, which are the main producers of active energy, FIG. 8 is not prepared to implement TDPA.

In order to explicitly extract the delayed communication block, the network representation of FIG. 8 is separated into two networks according to the type of the dependent source. FIG. 9 is a network representation of an i-th terminal that considers only a plurality of source ports, showing a network having only power sources. The delayed force sources are summed in series at the terminals so that each delayed force source is connected in series. Electrical branches connected in series share a common current. Thus, the power conjugate pair of each delayed effort source is determined as follows.

Where D _q is the amount of time-delay from the source terminal q to the output terminal i.

As a result, the power conjugate pair of the port source is identified at each delay. However, it is still not possible to identify the active energy from the communication time-delay. Here, a method is proposed for shifting the source to an un-delayed position and clarifying active energy components due to time-delay by adding a transport network (which may be referred to as a time delay power network (TDPN)). The TDPN makes it possible to estimate the amount of active energy due to time-delay by simply comparing the amount of energy at the delayed position with the original source of energy and active energy due to time-delay separation and non-delayed position . By applying TDPN for all delayed port sources, the network circuit of Fig. 9 can be transformed as shown in Fig. 10 is an enhanced network representation of an i < th > terminal with multiple receive sources.

FIG. 11 shows a network representation in which only the velocity sources are considered in FIG. Unlike the summation of the source ports, the summation of the flow sources shown in equation (16) can be expressed as a parallel circuit. The voltages across all the branches of the parallel circuits are the same, and therefore, the commitment power pair for each delayed flow source is as follows.

Where f _ci is the calculated output control force of the speed controller at the i-th terminal, terminal i, and D _p is the amount of time-delay from source terminal p to end terminal i.

Similarly, the TDPNs allow the flow source to be shifted to an un-delayed position and the TDPNs in between to identify the active energy components due to the time-delay. The electrical representation of FIG. 11 can be transformed as shown in FIG.

As a result, the general network representation of the decoupled subsystem for the i < th > terminal can be represented as in FIG. 13 by summing FIG. 10 and FIG. This provides a clear network representation that allows input and output energy to be estimated at each network port, including time-delay.

IV. How to Manipulate Transformed Networks Using TDPA

This section describes how to use TDPA to manipulate the translated network representation of the i-th terminal of a multiparty remote control system. Based on the basic nature of passivity, the entire network system can be passive and stabilized if all components of the network are passive. As shown in FIG. 13, most components such as the device Z _i and the velocity controllers K _p and K _d are passive so long as they are suitably designed. Ideally dependent sources (h _n (v _n ), g _m (f _m )) do not affect system passivity because all generated energy can be released as long as the system is passive. Only TDPNs can generate active energy and affect the passivity of the system.

The total energy stored in all TDPNs in the ith terminal is

Where E _v ^N and E _f ^N are the sum of the stored energy of all velocity commands and all force feedback in the TDPNs as follows:

According to Equation (6), the energy stored in each TDPN is estimated by subtracting the output energy from the input energy. Therefore, equation (21) can be written as follows.

Each TDPN has two energy flows from one port to another, and both ports can generate active energy. However, since a dependent source can absorb a finite amount of energy, the energy flow to that source does not affect the passivity of the TDPN. Therefore, considering only the energy flows from the source to the terminal is sufficient to ensure the passivity of the TDPN.

In FIG. 13, the network representation with TDPN has a clear power conjugate pair in each channel. As a result, for each type of source, the input output energies at each TDPN can be obtained as follows.

- For speed source:

- For force sources:

As described in Section II, the TDPN is passive if the energy stored in the TDPN is maintained at a positive value over time. Therefore, the passivity condition at the i-th terminal is as follows.

While all output energies in the terminal in Equation 25 can be observed in real time, the input energies occur with network delay and are therefore not observable at the same sampling time. In this regard, the passive condition can be met by comparing the delayed input energy to the current output energy. As a result, the passivity condition of Equation 25 can be satisfied if the total output energy at one port does not exceed the total delayed input energies from all sources, and can be expressed as follows.

Where D _p and D _q are the time delays in the communication channel from the p th terminal to the q th terminal to the i th terminal.

To satisfy Equation 26, a PC may be introduced to emit active energy flows at the TDPN. To limit the output energy to less than the delayed input energy at the output port of each TDPN, two different types of PCs can be designed according to the source type as follows.

- Admittance type: For TDPN with flow source template

Here, the energy difference is as follows.

- Impedance type: For TDPN with the input sources

Here, the energy difference is as follows.

Figure 14 shows the resultant animated architecture of the i-th terminal of a multipoint remote control system via TDPA.

On the other hand, in the technique described above, it may be too conservative to independently scale all TDPNs, since there may be some degree of energy tolerance and may be used in other channels. To minimize the amount of energy emitted and to make the framework less conservative, using a single PC for each source type, instead of introducing a PC for each TDPN, It may be sufficient to emit.

For terminals with flow sources, a single admittance type PC can be designed to be sufficient to drive this channel and to:

Here, the sum of the energies of the respective TDPNs having flow sources is given as follows.

Where E _in ^vp is the first input energy from the first type source p to the terminal i and E _out ^vp is the first output energy from the first type source p through the network to terminal i, T is the sampling time, _Dp Is the time delay in the transmission over the network to the terminal i in the first type of the source p, f _ci is a resultant force of the position controller in a terminal i.

For terminals with source ports, it is sufficient to drive this channel with a single impedance type PC and can be designed as follows.

Here, the sum of the energies of the respective TDPNs having the port sources is as follows.

As a result, the less conservatively manipulated i-th terminal via TDPA can be designed as shown in Fig. Figure 15 shows the i-th terminal that is less conservatively manipulated using TDPA. The input energies calculated at the non-delayed locations are sent to the terminal with delay. The transmitted energies with the same type of source are summed and represented as one delayed input energy. For example, energies transmitted from all flow sources are summed and represented as a single delayed input energy. The output energy is also calculated separately at the terminal for the source type. Finally, two PCs are introduced to regulate the respective output energy under delayed input energy.

The stabilization of the remote control between a plurality of terminals as described above may be implemented by a remote control stabilization system including electronic devices such as a computer program or a module with a built-in computer program. Here, at least some of the terminals of the plurality of terminals include at least one of a first type source (e.g., a flow source, a speed signal source) and a second type source (e.g., . The remote control stabilization system includes a network connecting a plurality of terminals and the network receives a signal (velocity signal, or force signal) from at least one of the first type source and the second type source to terminal i and transmits the signal to terminal i Lt; / RTI >

In addition, the remote control stabilization system may include a passivity observer for monitoring the passivity of the network, a passivity controller for consuming the PO and energy to maintain the passivity of the network, and a PC. The passivity observer has a first input energy from a first type source to a terminal i which is any of a plurality of terminals as shown in Figure 15, a second input energy from a second type source to a terminal i, Type source to the terminal i, and a second output energy output from the second type source through the network to the terminal i, respectively, to thereby measure the input energy and the output energy per source type, respectively Can be monitored.

At least one type of the first type source and the second type source may be composed of a plurality of sources and may be composed of a plurality of flow sources and a plurality of earphone sources as shown in Fig. A plurality of flow sources, and a plurality of flow sources.

The passivity observer includes a first observer observing a first input energy from flow sources to the network towards terminal i and a first output energy from the flow sources to the terminal i through the network and a second observer And a second observer for observing a second input energy to the network and a second output energy from the sources to the terminal via the network. The passivity observer is defined as the sum of the source types of input energy from a plurality of sources to terminal i, the sum of the source types of output energy output to terminal i through the network, and the total input energy and total output energy Can be monitored and calculated.

15 may include a first controller of an admittance type and a second controller of an impedance type and the first controller of an admittance type may include a plurality of flow sources connected in parallel and a plurality of flow sources connected between the network and the terminal i And a second controller of the impedance type can be interpreted as functioning between a plurality of input sources connected in series and their network and terminal i. The admittance type passivity controller can adjust the amount of energy output from the speed signal source, which is a flow source, to the terminal i via the network, using the damping factor? Defined by equation (31). The impedance type passivity controller can adjust the amount of energy output to the terminal i through the network from the force signal source, which is the port source, using the damping factor? Defined by equation (33).

V. Overall Passivity of the Multilateral Remote Control System

In the previous section, we have proposed a general framework to automate a single decoupled network of multiparty remote control systems. It is possible that similar frameworks can simulate different decoupled networks and show the passivity of all decoupled network systems, thereby showing the overall passivity of the multi-party remote control system.

Due to the network representation method introduced in Section III, in the form of a block diagram, a multiparty remote control system can be converted to a network representation with energy flows. Figure 16 shows the energy flows of a multipoint remote control system. There are input and output energy flows at each terminal that can be master or slave robots. A portion of the input energy is stored in the network and released from the network, and the remainder of the input energy is transmitted to the other terminals. The total stored amount of energy in the network is calculated by summing all the input energies and subtracting all the output energies of each terminal, and is expressed as:

Here, E _in ^M1 is the input energy from the master 1 to the network, and E _out ^S1 represents the output energy from the network to the slave 1. If the amount of energy stored in the network is maintained at 0 or more over the entire time, then the multipurpose remote control system can be said to be passive.

However, as described in equation (25), it is technically impossible to monitor E _stored at the same sampling time due to the distributed nature of the multipoint remote control system. Implementing TDPA in Equation 35 may seem simple. However, it was not clear to determine what input and output energy should be paired at the signal level to implement TDPA as discussed earlier in this disclosure.

Considering the contribution to the target terminal, the input energy can be separated as follows, taking into account contributions to the target terminal, such as separating the input forces and velocities as shown in equation (15).

Where E _in ^M1toS1 (k) is the input energy contribution of master 1 to slave 1, and E _in ^M1toS2 (k) is the input energy contribution of master 1 to slave 2. It is noted that, as discussed in Section III, the input energy contribution can be identified by finding the contribution of the input signals and its power conjugate pair.

Thus, E _stored can be decoupled to several subnetworks for the end-terminal as if it had decoupled the multipoint remote control system in Section III. For example, when considering slave 1, since the input energy contribution to slave 1 of other terminals can be identified as in Fig. 17 (a), that part of the stored energy of the network can be expressed as:

In addition, the energy stored with respect to master 2 as an end-terminal is expressed in a similar manner as follows.

In general, the stored energy of the subnetwork is the subtraction of the output energy from the sum of the input energy contributions from the other side of the terminal to the terminal.

As a result, the total energy of the network can be reformulated as the sum of the stored energy of each subnetwork as follows.

If each stored energy is held at a positive value over time, the entire system can be concluded to be passive.

In section IV, how to make the stored energy of each sub-network a positive value based on TDPA has already been discussed. Thus, it can be concluded that the proposed framework can guarantee the passivity of a multiparty remote control system.

VI. Experiment

The proposed framework was tested with a three - way remote control system. 18 shows a three-way remote control system having a phantom premium 6 DOF as a master 1, a phantom OMNI as a master 2, and a phantom premium as a slave. Trainer / trainee and group remote control scenarios were considered. Dual operators can remotely manipulate a single slave at the same time with different amounts of authority based on their experience level or control capability. The interactive force of the slave is fed back to both operators to transmit a sense of touch. Operators were coupled to each other by sharing the velocity signals from Master 1 to Master 2 and the control force signals from Master 2 to Master 1. Z _c and Z _sc are the local position controllers of Master 2 and Slave. A fixed communication time-delay of 100 msec was introduced in channels T ₁ , T ₂ and T ₅ , and a fixed communication time-delay of 150 msec was introduced in channels T ₃ , T ₄ and T ₆ . Note that the proposed framework can also cover time-varying delays.

In order to make the tripod remote control system passive, it first identifies the contribution of the input signals to each terminal as in equation (15). In the case of a slave end-terminal, there is only an output rate from the network and no output power is present. The output speed from the network to the slave consists of the contribution of the input speeds from both masters as follows.

Where alpha is a power factor that determines which master will have a greater effect on the slave. In this experiment, α was set to a value of 0.5.

On the other hand, in the case of Master 1, only the force output from the network without velocity output was as follows.

Where f _in ^M2 is the calculated force of the master 2 position controller for coupling both masters kinesthetically and f _in ^S from both masters is the position controller of the slave to track the target speed command It was the power of.

In the case of Master 2, the force and speed output from the network were simultaneously as follows.

Where f _in ^S was the calculated output force of the slave's position controller and v _in ^M1 was the velocity of master 1. Since the output force and velocity signals of each end-terminal are explicitly indicated along with the input signals from the other side of the network, a block diagram of a three-way remote control system is shown for each end- It can be converted into three network representations decoupled. Each representation includes TDPNs to clarify active energy components due to time-delay and POs / PCs for manipulating each TDPN. For further details, please refer to sections III and IV.

20 and 21 show experimental results when the proposed method is not used. When both operators moved their respective masters to remotely control a single slave (with a given rights factor), both masters and slaves began to vibrate in free space, and the system eventually deviated as in Figure 20 I abandoned it. As expected, the output energy exceeded the input energy in all TDPNs, as shown in FIG. However, when using the proposed method as designed in FIG. 19, both masters and slaves are able to maintain stable position tracking and force as shown in FIG. 22, even though operators have created inferior operating conditions by causing the slave to collide with the external environment. Feedback performance. As shown in FIG. 23, in all the TDPNs, the output energy was limited by the input energy.

As discussed in FIG. 15, instead of manually manipulating all TDPNs independently, introducing one PC for each source type makes the design less conservative. This possibility can be clearly seen in Fig. There were noticeable energy margins in the TDPN (from master 1 to master 2), TDPN 3 (from master 1 to slave) and other TDPNs such as TDPN 2 and TDPN 4, and there were also small energy margins. Figure 24 shows a less conservative version of the passive technique for Master 1 and Slave. It is sufficient to manually tune the TDPNs for each type of source with a single PC. The design for the Master 2 terminal was the same as in Figure 19, with only a single input for each type of source.

In these experiments, operators manipulated the slaves more strongly than previous experiments shown in FIG. In constrained and free motion, the proposed technique showed a stable position / force response. The output energy at each end-terminal is limited by the summed input energy as in FIG. For the Master 2 terminal, in this case, the summed input energy of the sources of each time and the summed output energy were compared. Unlike FIG. 23, since some energy margins in one TDPN were used for different TDPNs, there was almost no energy margins for the entire interaction region, thereby reducing unnecessary energy correction, resulting in a less conservative technique.

VII. Discussion and Conclusion

In the above, a general TDPA-based framework for stable multi-party remote control has been proposed, and a framework has been tested with a three-way remote control system. The design procedure of the proposed framework can be summarized as follows.

One) The output signals at the terminals are represented by a combination of the contributions of the input signals from the other side of the network.

2) Each terminal is converted to a network circuit having output signals as ports or flow sources depending thereon.

3) The port sources are connected in series and the delayed flow sources are connected in parallel.

4) Introduces a TDPN by shifting the port or flow source to an un-delayed position at each delay.

5) Introduce POs / PCs in each TDPN to tune the TDPN, or introduce one PO / PC for TDPNs with the same type of source.

A key aspect of these frameworks is how to convert a conventional block diagram of a multiparty remote control system into a network representation with definite energy flows for the implementation of TDPA. In the present disclosure, such a transformation has evolved without using specific system-dependent parameters such as master / slave dynamic parameters, controller gains, amount of time-delay, and the like. Thus, the greatest advantage of the proposed framework is that such a framework can be implemented on any multipart remote control system, regardless of the amount of time-delay, the multiprotocol control architecture, and the number of masters / slaves. For convenience of explanation, in the experimental example, the idea of the present disclosure was tested with a three-way remote control system with a constant time-delay, but this framework requires a larger number of masters / slaves and a multi- It can be implemented without difficulty on a steering system. Since these frameworks deal only with network channels, frameworks can be applied regardless of the nonlinear dynamics of the master and slave robots. Also, multi-DOF extensions may be possible by implementing the method of the present disclosure independently in each Cartesian coordinate system.

Those skilled in the art will also appreciate that the various illustrative logical blocks, modules, circuits, and algorithm steps herein may be implemented as electronic hardware, computer software, or combinations of both. To clearly illustrate the interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps have been described generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.

The various illustrative logical blocks, modules, and circuits described in connection with the present disclosure may be implemented or performed with a general purpose processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (PLD), discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to implement the described functions. A general purpose processor may be a microprocessor, but, in the alternative, the processor may be a commercial processor, controller, microcontroller, or state machine. A processor may be implemented as, for example, a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration.

The steps of a method or algorithm described in connection with the present invention may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. The software module may reside in a RAM memory, a flash memory, a ROM memory, an EPROM memory, an EEPROM memory, registers, a hard disk, a portable disk, a CD-ROM, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor such that the processor can read information from, and write information to, the storage medium. Alternatively, the storage medium may be integral to the processor. The processor and the storage medium may be located in an ASIC. The ASIC may be located at the user terminal. In the alternative, the processor and the storage medium may reside as discrete components in a user terminal.

In one or more exemplary embodiments, the functions described may be implemented in hardware, software, firmware, or a combination thereof. When implemented in software as a computer program product, the functions may be stored on or transmitted via one or more instructions or code on a computer-readable medium. Computer-readable media includes computer storage media and communication media including any medium for facilitating transfer of a computer program from one place to another. The storage medium may be any available media that can be accessed by a computer. For example, such a computer-readable medium may store or store program code means as required in the form of a RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, Or any other medium which can be used to communicate, and which can be accessed by a computer. In addition, any connection means may be considered as a computer-readable medium. For example, if the software is transmitted over a wireless technology such as a coaxial cable, a fiber optic cable, a twisted pair, a digital subscriber line (DSL), or infrared, radio, and microwave from a web site, server, Wireless technologies such as coaxial cable, fiber optic cable, twisted pair cable, DSL, or infrared, radio, and microwave may be included in the definition of such a medium. The discs and discs used here include compact discs (CDs), laser discs, optical discs, DVDs, floppy discs, and Blu-ray discs where disc plays the data magnetically, As shown in FIG. The combinations should also be included within the scope of computer readable media.

The embodiments of the present invention described above are disclosed for the purpose of illustration, and the present invention is not limited thereto. It will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the spirit and scope of the invention.

Claims (8)

A system for stabilizing remote control between a plurality of terminals,
Wherein at least some of the plurality of terminals include at least one of a first type source and a second type source,
The system comprises:
A network connecting the plurality of terminals, the network receiving signals from at least one of a first type source and a second type source to terminal i and transmitting to the terminal i;
A passivity observer for monitoring the passivity of the network; And
And a passivity controller that consumes energy such that the network maintains passivity,
Said passive observer having a first input energy directed from said first type source to said terminal i, a second input energy directed from said second type source to said terminal i, i and a second output energy output from the second type source through the network to the terminal i,
Wherein the passivity controller adjusts an amount of energy output to the terminal i based on the monitoring result,
A system for stabilizing remote control between a plurality of terminals. The method according to claim 1,
Wherein at least one of the first type source and the second type source comprises a plurality of sources,
Wherein the passivity observer comprises: a sum of source types of input energy directed from the plurality of sources to the terminal i, a sum of source types of output energy output to the terminal i through the network, and a sum To calculate the difference between the input energy and the total output energy,
A system for stabilizing remote control between a plurality of terminals. 3. The method according to claim 1 or 2,
The first type source generates a velocity signal,
Wherein the passivity controller comprises an admittance type passivity controller to adjust the amount of energy output from the first type source to the terminal i via the network using a damping factor?
A system for stabilizing remote control between a plurality of terminals.
≪ Formula (a) >

Here, E _i ^v (k)

E _in ^vp is the first input energy from the first type source p to the terminal i and E _out ^vp is the first output energy from the first type source p through the network to the terminal i , DELTA T is the sampling time, _Dp Is the time delay in the transmission over the network to the terminal i in the first type of the source p, f _ci is a resultant force of the position controller in a terminal i. 3. The method according to claim 1 or 2,
Said second type source generating a force signal,
Wherein the passivity controller comprises an impedance type passivity controller to adjust the amount of energy output from the second type source to the terminal i via the network using a damping factor?
A system for stabilizing remote control between a plurality of terminals.
Equation (b)

Here, E _i ^f (k)

E _in ^vp is the second input energy from the second type source q to the terminal i and E _out ^vp is the second output energy from the second type source q through the network to the terminal i , DELTA T is the sampling time, _Tq Is the time delay in transmission over the network from the second type source q to terminal i, and v _i is the resulting rate at terminal i. CLAIMS 1. A method for stabilizing remote control between a plurality of terminals,
Wherein the plurality of terminals are connected by a network and at least some of the plurality of terminals include at least one of a first type source and a second type source,
The method comprises:
A first input energy from the first type source to the terminal i, a second input energy from the second type source to the terminal i, and a second input energy from the first type source to the terminal i via a passivity observer monitoring the passivity of the network, Monitoring a first output energy output to the terminal i via the network and a second output energy output from the second type source to the terminal i via the network, respectively; And
And adjusting an amount of energy output to the terminal i through a passivity controller that consumes energy to maintain passivity based on the monitoring result.
A method for stabilizing remote control between a plurality of terminals. 6. The method of claim 5,
Wherein at least one of the first type source and the second type source comprises a plurality of sources,
Wherein the monitoring step comprises: a sum of source types of input energy directed from the plurality of sources to the terminal i, a sum of source types of output energies output to the terminal i through the network, Calculating a difference between a total input energy and a total output energy,
A method for stabilizing remote control between a plurality of terminals. The method according to claim 5 or 6,
The first type source generates a velocity signal,
Wherein the adjusting comprises adjusting the amount of energy output from the first type source to the terminal i via the network, using a damping factor? Defined by equation (a) via an admittance type passivity controller / RTI >
A method for stabilizing remote control between a plurality of terminals.
≪ Formula (a) >

Here, E _i ^v (k)

E _in ^vp is the first input energy from the first type source p to the terminal i and E _out ^vp is the first output energy from the first type source p to the terminal i through the network , DELTA T is the sampling time, _Dp Is the time delay in transmission over the network from the first type source p to terminal i and f _ci is the resulting position controller force at terminal i. The method according to claim 5 or 6,
Said second type source generating a force signal,
Wherein the adjusting comprises adjusting the amount of energy output from the second type source to the terminal i via the network using an impedance type passivity controller and using a damping factor? Including,
A method for stabilizing remote control between a plurality of terminals.
Equation (b)

Here, E _i ^f (k)

E _in ^vp is the second input energy from the second type source q to the terminal i and E _out ^vp is the second output energy from the second type source q through the network to the terminal i , DELTA T is the sampling time, _Tq Is the time delay in transmission over the network from the second type source q to terminal i, and v _i is the resulting rate at terminal i.

Images