WO2019039646A1 - Robot control system - Google Patents

Robot control system Download PDF

Info

Publication number
WO2019039646A1
WO2019039646A1 PCT/KR2017/010883 KR2017010883W WO2019039646A1 WO 2019039646 A1 WO2019039646 A1 WO 2019039646A1 KR 2017010883 W KR2017010883 W KR 2017010883W WO 2019039646 A1 WO2019039646 A1 WO 2019039646A1
Authority
WO
WIPO (PCT)
Prior art keywords
motion
external force
robot
adaptive
actuator
Prior art date
Application number
PCT/KR2017/010883
Other languages
French (fr)
Korean (ko)
Inventor
전동수
조창연
Original Assignee
주식회사 토룩
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 주식회사 토룩 filed Critical 주식회사 토룩
Publication of WO2019039646A1 publication Critical patent/WO2019039646A1/en

Links

Images

Classifications

    • BPERFORMING OPERATIONS; TRANSPORTING
    • B25HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25JMANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J17/00Joints
    • B25J17/02Wrist joints
    • B25J17/0208Compliance devices
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B25HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25JMANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J17/00Joints
    • B25J17/02Wrist joints
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B25HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25JMANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J9/00Programme-controlled manipulators
    • B25J9/10Programme-controlled manipulators characterised by positioning means for manipulator elements
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B25HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25JMANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J9/00Programme-controlled manipulators
    • B25J9/16Programme controls
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B25HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25JMANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J9/00Programme-controlled manipulators
    • B25J9/16Programme controls
    • B25J9/1628Programme controls characterised by the control loop
    • B25J9/1633Programme controls characterised by the control loop compliant, force, torque control, e.g. combined with position control

Definitions

  • the present invention relates to a robot control system, and more particularly, to a robot control system for controlling the operation of the actuator of a robot including at least one actuator, the robot control system being characterized in that the actuator is adapted to perform a first motion with a predetermined locus Motion engine; And an adaptive reaction algorithm that senses an external force applied to the robot and changes a locus of the first motion corresponding to the external force when an external force is applied to the robot, To a robot control system capable of realizing an interaction with a robot in a high level and an external environment.
  • the conventional robot has a limitation only as a mechanism for reproducing a program that has been previously planned and programmed in realizing a motion according to a specific purpose.
  • These limitations act as an impediment to human interaction. Therefore, it is difficult to achieve the goal of being a human closer to the robot, which is a future robot.
  • a pet animal When remembrance of a living animal, a pet animal, behaves according to a predetermined behavior of a pet, or an action of adapting by changing the trajectory or posture of an action when an external force is applied by a human hand touching the posture.
  • the present invention is conceived to solve the above problems, and it is an object of the present invention to provide a robot control system for controlling the operation of the actuator of a robot including at least one actuator, wherein the actuator performs a first motion having a predetermined locus A motion engine; And an adaptive reaction algorithm that senses an external force applied to the robot and changes a locus of the first motion corresponding to the external force when an external force is applied to the robot, And a robot control system in which the motion of the robot is deformed according to the robot control system, and the interaction between the robot and the external environment at a high level can be realized.
  • a robot control system for controlling the operation of the actuator of a robot including at least one actuator, the robot control system comprising: a first motion A motion engine to perform; And an adaptive reaction algorithm that senses an external force applied to the robot and changes the locus of the first motion corresponding to the external force when an external force is applied to the robot.
  • the robot control system includes: a sensor unit for sensing an external force; And an adaptive coefficient determining unit for determining an adaptive coefficient according to a predetermined adaptive coefficient determining algorithm for an external force sensed by the sensor unit, Thereby changing the locus of the first motion.
  • the actuator is constituted by a predetermined rotary actuator, and the compliant reaction algorithm determines an adaptive torque according to the following equation 1-A,
  • the amount of change of the first motion is determined according to Equation 1-B below.
  • the actuator is configured with a predetermined linear actuator, and the compliant reaction algorithm determines a compliance force according to Equation 2-A below,
  • FC adaptive torque
  • FS external force sensed by the sensor part
  • C adaptation coefficient
  • the amount of change of the first motion is determined according to the following equation (2-B).
  • the robot control system has a compliant reaction algorithm in which a motion engine that generates a first motion having a predetermined trajectory changes its trajectory of the first motion in accordance with an external force, It is possible to change the trajectory of the first motion and thus to realize the interaction with the external environment with the robot at the higher level.
  • FIG. 1 is a block diagram of a robot control system according to the present invention.
  • FIG. 2 is a diagram showing a change in the locus of motion of the actuator by the robot control system according to the present invention.
  • FIG. 3 is a diagram illustrating a relationship between a posture and an external force of the robot by the robot control system according to the present invention.
  • a robot control system for controlling the operation of an actuator of a robot comprising at least one actuator, the robot comprising: a motion engine for causing the actuator to perform a first motion with a predetermined trajectory; And an adaptive reaction algorithm that senses an external force applied to the robot and changes a locus of the first motion corresponding to the external force when an external force is applied to the robot, The robot's motion is modified in accordance with the robot, and the robot can interact with the external environment at a high level.
  • a robot control system (1) is a robot control system (1) for controlling the operation of the actuator of a robot including at least one actuator, characterized in that the actuator is adapted to perform a first motion with a predetermined locus A motion engine 100; And an adaptive reaction algorithm for changing the locus of the first motion corresponding to the external force when an external force is applied to the robot.
  • the motion engine is closer to the concept of a control device having an algorithm for generating an operation command for a predetermined actuator, rather than a concept as an engine as a mechanical device for generating power.
  • the motion engine 100 generates a predetermined signal so that an actuator provided in the robot performs a first motion having a predetermined trajectory.
  • the motion engine 100 may include a predetermined input unit and a control system.
  • the input unit may be a keypad, a panel, or the like capable of receiving various signals such as various key inputs inputted by a user.
  • the control system may be a predetermined CPU for generating an actuating signal for actuating an actuator provided in the robot by a signal inputted through an input unit.
  • contents of various first motions such as the locus of the first motion can be inputted and stored in advance, and the first motion can be performed by the actuator by the signal transmitted by the input unit.
  • the motion transformer 200 has an adaptive reaction algorithm.
  • the adaptive reaction algorithm alters the locus of the first motion corresponding to the external force when an external force is applied to the robot.
  • the adaptive reaction algorithm generates an adaptive motion having a predetermined amount of change that deforms the first motion in response to an external force.
  • the second motion is generated by combining the adaptive motion generated by the external force and the first motion generated by the command input through the input unit, and the actuator operates according to the locus of the second motion.
  • the robot control system may further include a sensor unit 300 for sensing an external force.
  • the sensor unit 300 can sense an external force applied to the robot.
  • the sensor unit 300 may include any sensor such as a sensor for sensing pressure of various load cells or an optical sensor. Therefore, the term “ external force " is not necessarily limited to " force ", but can be understood as a concept encompassing all the influences exerted on the robot in various external environments except for the input signal through the input unit.
  • the motion engine 100 determines the adaptation coefficient based on the external force sensed through the sensor unit 300.
  • the adaptation coefficient may be determined by substituting the external force sensed by the sensor unit 300 into a predetermined adaptation coefficient determination algorithm, and the determination of the adaptation coefficient may be performed by the predetermined adaptation coefficient determination unit 400.
  • the adaptation determination algorithm included in the adaptation coefficient determination unit 400 determines whether or not the adaptation algorithm is one of the three You can make a choice. For example, when the magnitude of the external force is small, a high adaptation coefficient is given, and when the magnitude of the external force is large, a low adaptation coefficient can be given. According to this, the conformity to the external force is relatively low, so it has a low adaptation coefficient, the conformity has the highest compliance coefficient with the external force, and the adaptation to the external force has the lowest compliance factor .
  • the adaptive reaction algorithm changes the trajectory of the first motion according to the adaptive coefficient determined in the adaptive coefficient determiner 400.
  • the adaptive reaction algorithm can change the trajectory of the first motion or change it to a very small size.
  • the amount of change of the first motion occurs in a small size, so that the degree of the actuator following the trajectory of the first motion becomes relatively high.
  • the adaptive coefficient determiner 400 selects to comply, the trajectory of the first motion generated by the motion engine 100 in response to the external force is deformed to a relatively large extent. Therefore, the degree of follow-up with respect to the original trajectory of the first motion becomes relatively low.
  • the trajectory of the first motion can be deformed in the direction opposite to the direction in which the external force comes. Accordingly, the deformed motion is greatly deformed in the trajectory of the first motion, and the degree of follow-up of the trajectory of the first motion may be very low.
  • the first motion is generated by selecting one of the motions embedded in the control system by the actuating signal inputted through the input unit, and the adaptive reaction is generated by the external force sensed through the sensor unit 300 A new motion can be generated by the algorithm by modifying the first motion.
  • the adaptive torque is calculated according to Equation 1-A as follows.
  • TC denotes an adaptive torque
  • TS denotes an external force measured by the sensor unit 300.
  • C represents the compliance coefficient (compliance coefficient).
  • the adaptation coefficient can be determined by substituting the measured external force into a predetermined adaptation coefficient determination algorithm.
  • the adaptive coefficient determination algorithm may include, for example, a predetermined emotion engine or the like. That is, when the emotion state of the robot is set to the current emotion state, the adaptation coefficient can be determined so that the compliance degree is increased. When the non-compliance degree is set, the compliance coefficient can be determined so that the compliance degree is decreased.
  • the adaptive torque is used to derive the amount of motion variation that changes the locus of the first motion according to Equation 2-A below.
  • is the amount of motion variation
  • KS is the elastic modulus of the system.
  • Equation 3-A the finally changed second motion is generated as follows.
  • the magnitude of the second motion is derived in the form of subtracting the amount of motion variation from the magnitude of the first motion, but the form of summing is also possible.
  • the actuator generates the first motion according to the setting of the motion engine 100, and the size of the first motion is SP1.
  • the SP1 may be, for example, a predetermined straight line distance.
  • the adaptive force is calculated as shown in Equation 1-B below.
  • FC is a conforming force
  • TS represents an external force measured by the sensor unit 300.
  • C represents the compliance coefficient (compliance coefficient).
  • the adaptation coefficient can be determined by substituting the measured external force into a predetermined adaptation coefficient determination algorithm.
  • ⁇ SP is the amount of motion variation
  • KA is the stiffness coefficient of the system.
  • Equation 3-B the finally changed second motion is generated as follows.
  • the size of the second motion is derived in the form of subtracting the amount of motion variation from the size of the first motion, but the form of summing is also possible.
  • the first motion by the motion engine 100 is not limited to the one involving movement.
  • the first motion by the motion engine 100 is an operation for maintaining a specific posture rather than a movement along a specific trajectory, the posture is maintained.
  • the adaptive reaction algorithm implements an appropriate response motion according to the external force. That is, the above-described amount of motion variation appears as a reaction motion.
  • the adaptive reaction algorithm takes a predetermined motion conforming thereto , And when the external force is removed, it returns to the original position and maintains the predetermined posture according to the first motion, which is the motion of maintaining the existing posture.
  • FIG. 3 An example of this is shown in FIG. First, it is assumed that a downward external force is applied to the head T of the robot as shown in Fig. 3 (a).
  • the motion engine 100 generates the first motion to maintain the attitude, but as the downward external force is sensed, the adaptive reaction algorithm generates a predetermined adaptive motion that moves the actuator locus downward. Accordingly, the head of the robot moves downward.
  • the adaptive reaction algorithm alters the locus of the first motion according to the changed external force.
  • the second motion engine 200 may detect a change in the magnitude of the external force and reduce or increase the amount of motion variation of the actuator. In this case, the posture maintaining motion by the first motion can be made larger.
  • the robot control system 1 can realize a motion in which the first basic motion is controlled in real time to naturally adapt to the external force.
  • the robot control system 1 according to the present invention is a robot control system 1 according to the present invention in which the motion engine 100 generating the first motion has an adaptive reaction algorithm so that the adaptive current motion is generated by modifying the first motion according to the external force, It is possible to implement an interaction with the robot in the external environment.
  • the first motion of this type is an operation pre-inputted in the motion engine 100, and is a motion selected by a user's input signal.
  • the actuators provided at the respective joints of the robot are controlled with a predetermined locus so as to follow the target position according to time.
  • a predetermined locus For example, in the neck joint consisting of three joints, when the command is given to look at 45 degrees in 30 seconds from the head in three seconds, the yaw actuator and the pitch actuator of the neck joint follow a short control time unit A motion value to be generated is generated, and control according to the generated motion value is performed.
  • the robot when an external force is applied to the robot while performing the first motion that is input in advance, the trajectory of the first motion is changed in real time in the adaptive reaction algorithm. Therefore, the robot can make an appropriate interaction with the external force.
  • the robot control system 1 stores only the data of the first motion without data input for other complex motions other than the first motion, and the reaction to the external force applied in real- By changing the trajectory of the first motion, it is possible to perform immediate reaction motion. In other words, when receiving an external force, it receives the command again and does not follow the motion having the new trajectory but changes the trajectory of the first motion generated in the real time. Therefore, a small amount of calculation is required, . In addition, when the external force is removed, the motion of the first motion can be restored and the first motion can be performed again.

Landscapes

  • Engineering & Computer Science (AREA)
  • Robotics (AREA)
  • Mechanical Engineering (AREA)
  • Manipulator (AREA)

Abstract

The present invention relates to a robot control system and, more specifically, to a robot control system for controlling the operation of at least one actuator included in a robot, the system comprising: a motion engine for making the actuator carry out a first motion having a predetermined trajectory; and a motion change part having an adaptive response algorithm in which, when an external force is applied to the robot, the external force applied to the robot is sensed and the trajectory of the first motion is changed in response to the external force, wherein the motion of the robot can be changed in response to the external force applied to the robot, thereby implementing a high level of interaction between the robot and an external environment.

Description

로봇 제어 시스템Robot control system
본 발명은 로봇 제어 시스템에 관한 것으로서, 보다 상세하게는, 하나 이상의 액츄에이터를 포함하는 로봇의 상기 액츄에이터의 작동을 제어하는 로봇 제어 시스템으로서, 상기 액츄에이터가 소정의 궤적을 갖는 제1 모션을 수행하도록 하는 모션 엔진; 및 상기 로봇에 외력이 가해질 경우, 상기 로봇에 가해진 외력을 센싱하여 상기 외력에 대응하여 상기 제1 모션의 궤적을 변경시키는 순응형 반응 알고리즘을 갖는 모션 변형부;를 포함하여, 로봇에 가해지는 외력에 따라서 로봇의 모션이 변형되어, 높은 단계의 로봇과 외부 환경에 대한 상호작용을 구현할 수 있는 로봇 제어 시스템에 관한 것이다.The present invention relates to a robot control system, and more particularly, to a robot control system for controlling the operation of the actuator of a robot including at least one actuator, the robot control system being characterized in that the actuator is adapted to perform a first motion with a predetermined locus Motion engine; And an adaptive reaction algorithm that senses an external force applied to the robot and changes a locus of the first motion corresponding to the external force when an external force is applied to the robot, To a robot control system capable of realizing an interaction with a robot in a high level and an external environment.
최근 로봇에 대한 관심도가 높아지고 많은 발전을 이루고 있다. 그러나, 종래의 로봇은 특정 목적에 따라서 움직임을 구현하는 데 있어서 미리 기획되어 프로그래밍 된 행동을 재생하는 기계장치로서의 의미만을 갖는 한계를 갖고 있었다. 이러한 한계는 특히 서비스 또는 엔터테인먼트 분야에서 사용되는 로봇의 경우 인간과의 상호작용을 저해시키는 요소로 작용하게 된다. 따라서, 미래의 로봇이 지향하는 바인, 보다 인간에 가까운 로봇이라는 목표 달성을 어렵게 한다.Recently, interest in robots has increased and many improvements have been made. However, the conventional robot has a limitation only as a mechanism for reproducing a program that has been previously planned and programmed in realizing a motion according to a specific purpose. These limitations, in particular in the case of robots used in the field of services or entertainment, act as an impediment to human interaction. Therefore, it is difficult to achieve the goal of being a human closer to the robot, which is a future robot.
특히, 사람과의 접촉에 대해 종래의 기술은 많은 경우에 터치를 기반으로 하는 on/off 제어 정도로 그 이용이 국한되어 있다. 아울러, 힘 센서를 이용하여 외부 접촉 자극의 종류를 판별하는 경우에도, 그 외부 자극에 대한 순응 여부 및 순응 여부에 따른 적절한 대응 행동을 제공하지 않는다. 이러한 것은 로봇을 대할 때 살아있는 생명체처럼 느끼도록 하기 어렵고, 사람과 로봇 간의 상호작용을 방해하는 요인으로 작용한다.In particular, conventional techniques for contact with humans are limited in their use to the degree of touch-based on / off control in many cases. Further, even when the type of the external contact stimulus is discriminated by using the force sensor, it does not provide an appropriate response according to whether the external stimulus is compliant or not. It is difficult to feel like living creatures when handling robots, and it interferes with the interaction between human and robot.
살아있는 생물체인 애완용 동물을 연상해보면, 애완용 동물 소정의 동작, 또는 자세를 취하는 도중 사람의 손이 접촉하여 외력이 가해질 경우 동작의 궤적, 또는 자세를 변경시켜 순응하는 행동을 보인다.When remembrance of a living animal, a pet animal, behaves according to a predetermined behavior of a pet, or an action of adapting by changing the trajectory or posture of an action when an external force is applied by a human hand touching the posture.
따라서, 살아있는 생명체와 보다 유사한 동작을 보이는 로봇을 구현하기 위해서는, 외력을 감지하고, 감지된 외력에 따라서 동작 및 자세를 실시간으로 변동시키는 제어 시스템이 제공될 필요가 있다.Therefore, in order to implement a robot that exhibits a motion similar to that of a living creature, it is necessary to provide a control system that senses an external force and changes an operation and a posture in real time according to the sensed external force.
본 발명은 전술한 문제점을 해결하기 위해 안출된 것으로서, 본 발명은 하나 이상의 액츄에이터를 포함하는 로봇의 상기 액츄에이터의 작동을 제어하는 로봇 제어 시스템으로서, 상기 액츄에이터가 소정의 궤적을 갖는 제1 모션을 수행하도록 하는 모션 엔진; 및 상기 로봇에 외력이 가해질 경우, 상기 로봇에 가해진 외력을 센싱하여 상기 외력에 대응하여 상기 제1 모션의 궤적을 변경시키는 순응형 반응 알고리즘을 갖는 모션 변형부;를 포함하여, 로봇에 가해지는 외력에 따라서 로봇의 모션이 변형되어, 높은 단계의 로봇과 외부 환경에 대한 상호작용을 구현할 수 있는 로봇 제어 시스템을 제공하는데 목적이 있다. SUMMARY OF THE INVENTION The present invention is conceived to solve the above problems, and it is an object of the present invention to provide a robot control system for controlling the operation of the actuator of a robot including at least one actuator, wherein the actuator performs a first motion having a predetermined locus A motion engine; And an adaptive reaction algorithm that senses an external force applied to the robot and changes a locus of the first motion corresponding to the external force when an external force is applied to the robot, And a robot control system in which the motion of the robot is deformed according to the robot control system, and the interaction between the robot and the external environment at a high level can be realized.
상술한 목적을 달성하기 위하여, 본 발명에 따른 로봇 제어 시스템은, 하나 이상의 액츄에이터를 포함하는 로봇의 상기 액츄에이터의 작동을 제어하는 로봇 제어 시스템에 있어서, 상기 액츄에이터가 소정의 궤적을 갖는 제1 모션을 수행하도록 하는 모션 엔진; 및 상기 로봇에 외력이 가해질 경우, 상기 로봇에 가해진 외력을 센싱하여 상기 외력에 대응하여 상기 제1 모션의 궤적을 변경시키는 순응형 반응 알고리즘을 갖는 모션 변형부;를 포함한다.In order to achieve the above object, a robot control system according to the present invention is a robot control system for controlling the operation of the actuator of a robot including at least one actuator, the robot control system comprising: a first motion A motion engine to perform; And an adaptive reaction algorithm that senses an external force applied to the robot and changes the locus of the first motion corresponding to the external force when an external force is applied to the robot.
바람직하게는, 상기 로봇 제어 시스템은 외력을 센싱하는 센서부; 상기 센서부에서 센싱된 외력에 대해 소정의 순응 계수 결정 알고리즘에 따라서 순응 계수를 결정하는 순응 계수 결정부;를 더 포함하며, 상기 순응형 반응 알고리즘은, 순응 계수 결정부에서 결정된 순응 계수에 따라서 상기 제1 모션의 궤적을 변경시킨다.Preferably, the robot control system includes: a sensor unit for sensing an external force; And an adaptive coefficient determining unit for determining an adaptive coefficient according to a predetermined adaptive coefficient determining algorithm for an external force sensed by the sensor unit, Thereby changing the locus of the first motion.
바람직하게는, 상기 액츄에이터는 소정의 회전 구동기로 구성되며, 상기 순응형 반응 알고리즘은, 하기 식 1-A 에 따라서 순응 토크를 결정하며, Preferably, the actuator is constituted by a predetermined rotary actuator, and the compliant reaction algorithm determines an adaptive torque according to the following equation 1-A,
TC = TS × CTC = TS x C
(식 1-A)(Formula 1-A)
(TC : 순응 토크, TS: 센서부에서 감지된 외력, C : 순응 계수)(TC: adaptive torque, TS: external force sensed by the sensor, C: adaptation coefficient)
하기 식 1-B 에 따라서 상기 제1 모션의 변화량을 결정한다.The amount of change of the first motion is determined according to Equation 1-B below.
Δθ = (TS-TC)/KS?? = (TS-TC) / KS
(식 1-B) (Formula 1-B)
(Δθ : 제1 모션의 변화량, KS : 시스템의 강성 계수)(??: change amount of the first motion, KS: stiffness coefficient of the system)
바람직하게는, 상기 액츄에이터는 소정의 직선 구동기로 구성되며, 상기 순응형 반응 알고리즘은, 하기 식 2-A 에 따라서 순응 힘을 결정하며,Preferably, the actuator is configured with a predetermined linear actuator, and the compliant reaction algorithm determines a compliance force according to Equation 2-A below,
FC = FS × C FC = FS × C
(식 2-A)(Formula 2-A)
(FC : 순응 토크, FS: 센서부에서 감지된 외력, C : 순응 계수)(FC: adaptive torque, FS: external force sensed by the sensor part, C: adaptation coefficient)
하기 식 2-B 에 따라서 상기 제1 모션의 변화량을 결정한다.The amount of change of the first motion is determined according to the following equation (2-B).
ΔSP = (FS-FC)/KA? SP = (FS-FC) / KA
(식 2-B) (Formula 2-B)
(ΔSP : 제1 모션의 변화량, KA : 시스템의 강성 계수)(? SP: change amount of the first motion, KA: stiffness coefficient of the system)
본 발명에 따른 로봇 제어 시스템은, 소정의 궤적을 갖는 제1 모션을 생성하는 모션 엔진이 외력에 대해 순응하여 상기 제1 모션의 궤적을 변경시키는 순응형 반응 알고리즘을 가짐으로써, 외력에 대해 순응하며 제1 모션의 궤적을 변경시킬 수 있고, 따라서 높은 단계의 로봇과 외부 환경에 대한 상호작용을 구현할 수 있다. The robot control system according to the present invention has a compliant reaction algorithm in which a motion engine that generates a first motion having a predetermined trajectory changes its trajectory of the first motion in accordance with an external force, It is possible to change the trajectory of the first motion and thus to realize the interaction with the external environment with the robot at the higher level.
도 1 은 본 발명에 따른 로봇 제어 시스템의 구조를 나타낸 도면이다.1 is a block diagram of a robot control system according to the present invention.
도 2 는 본 발명에 따른 로봇 제어 시스템에 의한 액츄에이터의 모션의 궤적의 변화를 나타낸 도면이다.FIG. 2 is a diagram showing a change in the locus of motion of the actuator by the robot control system according to the present invention.
도 3 은 본 발명에 따른 로봇 제어 시스템에 의한 로봇의 자세와 외력 간의 관계를 나타낸 도면이다.3 is a diagram illustrating a relationship between a posture and an external force of the robot by the robot control system according to the present invention.
본 발명은 하나 이상의 액츄에이터를 포함하는 로봇의 상기 액츄에이터의 작동을 제어하는 로봇 제어 시스템으로서, 상기 액츄에이터가 소정의 궤적을 갖는 제1 모션을 수행하도록 하는 모션 엔진; 및 상기 로봇에 외력이 가해질 경우, 상기 로봇에 가해진 외력을 센싱하여 상기 외력에 대응하여 상기 제1 모션의 궤적을 변경시키는 순응형 반응 알고리즘을 갖는 모션 변형부;를 포함하여, 로봇에 가해지는 외력에 따라서 로봇의 모션이 변형되어, 높은 단계의 로봇과 외부 환경에 대한 상호작용을 구현할 수 있다.A robot control system for controlling the operation of an actuator of a robot comprising at least one actuator, the robot comprising: a motion engine for causing the actuator to perform a first motion with a predetermined trajectory; And an adaptive reaction algorithm that senses an external force applied to the robot and changes a locus of the first motion corresponding to the external force when an external force is applied to the robot, The robot's motion is modified in accordance with the robot, and the robot can interact with the external environment at a high level.
이하, 첨부된 도면을 참조하여, 본 발명에 따른 바람직한 실시예에 대하여 설명한다. 본 실시예는 제한적인 것으로 의도된 것이 아니다.Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings. The present embodiments are not intended to be limiting.
본 발명에 따른 로봇 제어 시스템(1)은, 하나 이상의 액츄에이터를 포함하는 로봇의 상기 액츄에이터의 작동을 제어하는 로봇 제어 시스템(1)으로서, 상기 액츄에이터가 소정의 궤적을 갖는 제1 모션을 수행하도록 하는 모션 엔진(100); 및, 상기 로봇에 외력이 가해질 경우, 상기 외력에 대응하여 상기 제1 모션의 궤적을 변경시키는 순응형 반응 알고리즘을 갖는 모션 변형부(200)를 포함한다.A robot control system (1) according to the present invention is a robot control system (1) for controlling the operation of the actuator of a robot including at least one actuator, characterized in that the actuator is adapted to perform a first motion with a predetermined locus A motion engine 100; And an adaptive reaction algorithm for changing the locus of the first motion corresponding to the external force when an external force is applied to the robot.
여기서, 모션 엔진이라 함은 동력을 발생시키는 기계적 장치로서의 엔진으로서의 개념보다, 소정의 엑츄에이터에 대해 동작 명령을 발생시키는 알고리즘이 내장된 제어 장치의 개념에 가깝다.Here, the motion engine is closer to the concept of a control device having an algorithm for generating an operation command for a predetermined actuator, rather than a concept as an engine as a mechanical device for generating power.
모션 엔진(100)은 로봇에 구비된 액츄에이터가 소정의 궤적을 갖는 제1 모션을 수행하도록 소정의 신호를 발생시킨다. 이에 따라서, 모션 엔진(100)은 소정의 입력부와 제어 시스템을 포함할 수 있다. 입력부는 예컨대 사용자에 의해서 입력되는 각종 키 입력 등의 각종 신호를 수신할 수 있는 키패드, 패널 등일 수 있다. 제어 시스템은 입력부를 통해 입력된 신호에 의해서 로봇에 구비된 액츄에이터를 작동시키는 작동 신호를 생성하는 소정의 CPU 일 수 있다. 제어 시스템에는 제1 모션의 궤적 등과 같은 각종 제1 모션의 내용이 미리 입력되어 저장될 수 있으며, 입력부에 의해서 전달된 신호에 의해서 상기 제1 모션이 액츄에이터에 의해서 수행될 수 있다.The motion engine 100 generates a predetermined signal so that an actuator provided in the robot performs a first motion having a predetermined trajectory. Accordingly, the motion engine 100 may include a predetermined input unit and a control system. The input unit may be a keypad, a panel, or the like capable of receiving various signals such as various key inputs inputted by a user. The control system may be a predetermined CPU for generating an actuating signal for actuating an actuator provided in the robot by a signal inputted through an input unit. In the control system, contents of various first motions such as the locus of the first motion can be inputted and stored in advance, and the first motion can be performed by the actuator by the signal transmitted by the input unit.
모션 변형부(200)는 순응형 반응 알고리즘을 갖는다. 순응형 반응 알고리즘은 로봇에 대해서 외력이 가해질 경우, 외력에 대응하여 제1 모션의 궤적을 변경시킨다. The motion transformer 200 has an adaptive reaction algorithm. The adaptive reaction algorithm alters the locus of the first motion corresponding to the external force when an external force is applied to the robot.
이를 달리 설명하면, 순응형 반응 알고리즘은 외력에 대응하여 제1 모션을 변형시키는 소정의 변화량을 갖는 순응형 모션을 생성한다. 이와 같이 외력에 의해서 발생한 순응형 모션과, 입력부를 통해 입력된 명령에 의해서 발생한 제1 모션이 합쳐짐으로써 제2 모션이 발생하여, 액츄에이터는 상기 제2 모션의 궤적에 따라서 동작하게 된다. In other words, the adaptive reaction algorithm generates an adaptive motion having a predetermined amount of change that deforms the first motion in response to an external force. The second motion is generated by combining the adaptive motion generated by the external force and the first motion generated by the command input through the input unit, and the actuator operates according to the locus of the second motion.
예컨대, 도 2 에 도시된 바에 따라서 설명하면, 먼저 (a) 와 같이 상 방향으로 움직이는 제1 모션이 있다고 가정한다. 이에 대해 (b) 와 같이 측방향으로부터 외력이 가해질 경우, 상기 순응형 반응 알고리즘에 의해서, 액츄에이터는 외력의 방향에 순응하여 제1 모션의 궤적을 변경시키며, 액츄에이터는 변경된 궤적에 따라서 이동하게 된다. 이어서, (c) 와 같이 상기 외력이 제거되거나 약해지면 제1 모션의 원래 궤적으로 돌아오는 형태의 모션이 수행되게 되며, 최종적으로는 (d) 와 같이 제1 모션의 원래 궤적으로 복귀하게 된다.For example, as shown in FIG. 2, it is assumed that there is a first motion that moves upward in the direction (a). If an external force is applied from the lateral direction as shown in (b), the actuator changes the locus of the first motion according to the direction of the external force by the adaptive reaction algorithm, and the actuator moves according to the changed locus. Then, as shown in (c), when the external force is removed or weakened, a motion of returning to the original trajectory of the first motion is performed, and finally the original trajectory of the first motion is returned as shown in (d).
로봇 제어 시스템은, 외력을 센싱하기 위한 센서부(300)를 더 포함할 수 있다. 센서부(300)는 로봇에 가해지는 외력을 센싱할 수 있는 장치로서, 예컨대 각종 로드셀 등의 압력을 감지하는 센서, 또는 광 센서 등 어떠한 센서도 포함할 수 있다. 따라서, "외력"이라 함은 반드시 "힘"에 국한된 개념이 아니라 입력부를 통한 입력 신호를 제외한 각종 외부 환경에서 로봇에 가해지는 영향을 모두 포괄하는 개념으로 이해될 수 있다.The robot control system may further include a sensor unit 300 for sensing an external force. The sensor unit 300 can sense an external force applied to the robot. For example, the sensor unit 300 may include any sensor such as a sensor for sensing pressure of various load cells or an optical sensor. Therefore, the term " external force " is not necessarily limited to " force ", but can be understood as a concept encompassing all the influences exerted on the robot in various external environments except for the input signal through the input unit.
또한, 모션 엔진(100)은 상기 센서부(300)를 통해 감지된 외력을 토대로 순응 계수를 결정한다. 상기 순응 계수는 상기 센서부(300)에서 센싱된 외력을 소정의 순응 계수 결정 알고리즘에 대입하여 결정할 수 있으며, 이러한 순응 계수의 결정은 소정의 순응 계수 결정부(400)에서 수행될 수 있다. In addition, the motion engine 100 determines the adaptation coefficient based on the external force sensed through the sensor unit 300. The adaptation coefficient may be determined by substituting the external force sensed by the sensor unit 300 into a predetermined adaptation coefficient determination algorithm, and the determination of the adaptation coefficient may be performed by the predetermined adaptation coefficient determination unit 400. [
일 예로, 순응 계수 결정부(400)에 포함된 순응도 결정 알고리즘은, 센서부(300)를 통해 외력이 감지되었을 때, 외력의 크기에 따라서 버티기, 순응하기, 회피하기의 3 가지 중 어느 하나의 선택을 내릴 수 있다. 예컨대, 외력의 크기가 작을 경우 높은 순응 계수를 부여하며, 외력의 크기가 커질수록 낮은 순응 계수를 부여할 수 있다. 이에 따르면, 버티기는 외력에 대한 순응도가 비교적 낮아서 낮은 순응 계수를 갖고, 순응하기는 외력에 대한 순응도가 가장 높아서 높은 순응 계수를 가지며, 회피하기는 외력에 대한 순응도가 가장 낮아서 매우 낮은 순응 계수를 가질 수 있다.For example, when the external force is sensed through the sensor unit 300, the adaptation determination algorithm included in the adaptation coefficient determination unit 400 determines whether or not the adaptation algorithm is one of the three You can make a choice. For example, when the magnitude of the external force is small, a high adaptation coefficient is given, and when the magnitude of the external force is large, a low adaptation coefficient can be given. According to this, the conformity to the external force is relatively low, so it has a low adaptation coefficient, the conformity has the highest compliance coefficient with the external force, and the adaptation to the external force has the lowest compliance factor .
순응형 반응 알고리즘은 상기 순응 계수 결정부(400)에서 결정된 순응 계수에 따라서 제1 모션의 궤적을 변경한다. The adaptive reaction algorithm changes the trajectory of the first motion according to the adaptive coefficient determined in the adaptive coefficient determiner 400. [
예컨대, 순응 계수 결정부(400)에서 버티기를 선택하였을 경우, 순응형 반응 알고리즘은 제1 모션의 궤적을 변경하지 않거나, 또는 매우 작은 크기로 변경할 수 있다. 이와 같은 버티기 모션에서는 제1 모션의 변화량은 작은 크기로 발생하며, 따라서 액츄에이터는 제1 모션의 궤적을 추종하는 정도가 비교적 높게 된다. For example, when the adaptation coefficient determination unit 400 selects the obstacle, the adaptive reaction algorithm can change the trajectory of the first motion or change it to a very small size. In such a sticking motion, the amount of change of the first motion occurs in a small size, so that the degree of the actuator following the trajectory of the first motion becomes relatively high.
다음으로, 순응 계수 결정부(400)에서 순응하기를 선택하였을 경우, 외력에 반응하여 모션 엔진(100)에서 생성된 제1 모션의 궤적을 비교적 큰 폭으로 변형시키게 된다. 따라서, 제1 모션의 원래 궤적에 대한 추종 정도가 비교적 낮게 된다. Next, when the adaptive coefficient determiner 400 selects to comply, the trajectory of the first motion generated by the motion engine 100 in response to the external force is deformed to a relatively large extent. Therefore, the degree of follow-up with respect to the original trajectory of the first motion becomes relatively low.
이어서, 순응 계수 결정부(400)에서 회피하기를 선택하였을 경우, 외력이 오는 방향에 대해서 반대 방향으로 제1 모션의 궤적을 변형시킬 수 있다. 이에 따라서 변형된 모션은 제1 모션의 궤적에서 큰 폭으로 변형되며, 제1 모션의 궤적의 추종 정도는 매우 낮을 수 있다. Next, when avoidance is selected in the adaptive coefficient determination unit 400, the trajectory of the first motion can be deformed in the direction opposite to the direction in which the external force comes. Accordingly, the deformed motion is greatly deformed in the trajectory of the first motion, and the degree of follow-up of the trajectory of the first motion may be very low.
위에서 설명한 바와 같이, 제1 모션은 입력부를 통해 입력된 작동 신호에 의해서 제어 시스템 내에 내장된 모션 중 어느 하나의 모션을 선택함으로써 발생하며, 센서부(300)를 통해 감지된 외력에 의해서 순응형 반응 알고리즘이 상기 제1 모션을 변형시킴으로써 새로운 모션이 생성될 수 있다. As described above, the first motion is generated by selecting one of the motions embedded in the control system by the actuating signal inputted through the input unit, and the adaptive reaction is generated by the external force sensed through the sensor unit 300 A new motion can be generated by the algorithm by modifying the first motion.
본 발명에 따른 로봇 제어 시스템(1)의 상기 순응형 반응 알고리즘에 대해 상세히 설명하면 아래와 같다. The adaptive reaction algorithm of the robot control system 1 according to the present invention will be described in detail as follows.
먼저, 일 예로, 소정의 회전을 수행하는 회전 구동 액츄에이터의 경우에 대해서 설명한다. 이에 따라서, 모션 엔진(100)에서는 소정의 토크를 발생시켜 액츄에이터가 θ1 만큼의 회전 각도 값을 갖는 궤적의 제1 모션을 생성한다고 한다.First, as an example, the case of a rotary drive actuator that performs a predetermined rotation will be described. Accordingly, in the motion engine 100, it is assumed that a predetermined torque is generated so that the actuator generates the first motion of the locus having the rotation angle value of? 1.
센서부(300)에서 소정의 외력이 감지되면 먼저 아래와 같이 식 1-A 에 따라서 순응 토크를 계산한다. When a predetermined external force is sensed by the sensor unit 300, the adaptive torque is calculated according to Equation 1-A as follows.
TC = TS × CTC = TS x C
(식 1-A)(Formula 1-A)
여기서, TC 는 순응 토크이며, TS 는 센서부(300)에 의해서 측정된 외력을 나타낸다. C 는 순응 계수(컴플라이언스 계수)를 나타낸다. 앞서 설명한 바와 같이, 순응 계수는 측정된 외력을 소정의 순응 계수 결정 알고리즘에 대입하여 결정될 수 있다. 이러한 순응 계수 결정 알고리즘은 예컨대 소정의 감성 엔진 등을 포함할 수 있다. 즉, 로봇의 감성 엔진에서 현재 감성 상태를 "순응" 으로 설정한 경우, 순응도가 높아지도록 순응 계수를 결정할 수 있고, "비순응"으로 설정한 경우 순응도가 낮아지도록 순응 계수를 결정할 수 있다.Here, TC denotes an adaptive torque, and TS denotes an external force measured by the sensor unit 300. C represents the compliance coefficient (compliance coefficient). As described above, the adaptation coefficient can be determined by substituting the measured external force into a predetermined adaptation coefficient determination algorithm. The adaptive coefficient determination algorithm may include, for example, a predetermined emotion engine or the like. That is, when the emotion state of the robot is set to the current emotion state, the adaptation coefficient can be determined so that the compliance degree is increased. When the non-compliance degree is set, the compliance coefficient can be determined so that the compliance degree is decreased.
이어서, 상기 순응 토크를 이용하여 아래 식 2-A 에 따라서 제1 모션의 궤적을 변경시키는 모션 변화량을 도출한다. Δθ 는 모션 변화량이며, KS 는 시스템의 탄성 계수이다.Subsequently, the adaptive torque is used to derive the amount of motion variation that changes the locus of the first motion according to Equation 2-A below. Δθ is the amount of motion variation, and KS is the elastic modulus of the system.
Δθ = (TS-TC)/KS?? = (TS-TC) / KS
(식 2-A)(Formula 2-A)
상기와 같이 변화량만큼 제1 모션의 궤적이 변경된다. 이를 식으로 표현하면 아래 식 3-A 과 같다. 식 3-A 에 따라서, 최종적으로 변경된 제2 모션이 아래와 같이 생성된다. 식 3-A 에서는 제1 모션의 크기에서 모션 변화량을 감산하는 형태로 제2 모션의 크기가 도출되었으나, 합산하는 형태도 가능하다. The locus of the first motion is changed by the amount of change as described above. This expression is expressed by the following expression 3-A. According to Equation 3-A, the finally changed second motion is generated as follows. In Equation 3-A, the magnitude of the second motion is derived in the form of subtracting the amount of motion variation from the magnitude of the first motion, but the form of summing is also possible.
θ2 = θ1 - Δθ = θ1 - (TS-TC)/KSθ2 = θ1 - Δθ = θ1 - (TS-TC) / KS
(식 3-A)(Formula 3-A)
이를 직선 방향 모션을 수행하는 직선 구동 액츄에이터에 적용하여 설명하면, 아래와 같을 수 있다. This can be applied to a linear drive actuator that performs linear motion, as follows.
먼저, 모션 엔진(100)의 설정에 따라서 액츄에이터가 제1 모션을 발생시킨다고 하며, 제1 모션의 크기는 SP1 이라고 한다. SP1 은 예컨대 소정의 직선 거리일 수 있다.First, the actuator generates the first motion according to the setting of the motion engine 100, and the size of the first motion is SP1. The SP1 may be, for example, a predetermined straight line distance.
센서부(300)에서 소정의 외력이 감지되면 아래 식 1-B와 같이 순응 힘을 계산한다.When a predetermined external force is sensed by the sensor unit 300, the adaptive force is calculated as shown in Equation 1-B below.
FC = FS × CFC = FS × C
(식 1-B)(Formula 1-B)
여기서, FC 는 순응 힘이며, TS 는 센서부(300)에 의해서 측정된 외력을 나타낸다. C 는 순응 계수(컴플라이언스 계수)를 나타낸다. 앞서 설명한 바와 같이, 순응 계수는 측정된 외력을 소정의 순응 계수 결정 알고리즘에 대입하여 결정될 수 있다. Here, FC is a conforming force, and TS represents an external force measured by the sensor unit 300. C represents the compliance coefficient (compliance coefficient). As described above, the adaptation coefficient can be determined by substituting the measured external force into a predetermined adaptation coefficient determination algorithm.
이어서, 상기 순응 힘을 이용하여 아래 식 2-B 에 따라서 제2 모션의 궤적을 변경시키는 모션 변화량을 도출한다. ΔSP 는 모션 변화량이며, KA 는 시스템의 강성 계수이다.Subsequently, the adaptive force is used to derive the amount of motion variation that changes the trajectory of the second motion according to Equation 2-B below. ΔSP is the amount of motion variation, and KA is the stiffness coefficient of the system.
ΔSP = (FS-FC)/KA? SP = (FS-FC) / KA
(식 2-B) (Formula 2-B)
상기와 같이 모션 변화량만큼 제1 모션의 궤적이 변경된다. 이를 식으로 표현하면 아래 식 3-B 와 같다. 식 3-B 에 따라서, 최종적으로 변경된 제2 모션이 아래와 같이 생성된다. 식 3-B 에서는 제1 모션의 크기에서 모션 변화량을 감산하는 형태로 제2 모션의 크기가 도출되었으나, 합산하는 형태도 가능하다. The locus of the first motion is changed by the amount of the motion variation as described above. This expression is expressed by the following expression 3-B. According to Equation 3-B, the finally changed second motion is generated as follows. In Equation 3-B, the size of the second motion is derived in the form of subtracting the amount of motion variation from the size of the first motion, but the form of summing is also possible.
SP2 = SP1 - ΔSPSP2 = SP1 -? SP
(식 3-B)(Formula 3-B)
위 설명에서는 토크와 힘으로 각각 구분하여 설명하였으나, 이에 반드시 한정되는 것은 아니며, 어떠한 형태의 외력에 대해서도 적용될 수 있다.In the above description, the torque and the force are separately described. However, the present invention is not limited thereto, and can be applied to any external force.
한편, 상기 모션 엔진(100)에 의한 제1 모션은 반드시 움직임을 수반하는 것에 한정하지 않는다. 즉, 예컨대 모션 엔진(100)에 의한 제1 모션이 특정 궤적을 따른 움직임이 아닌 특정한 자세를 유지하도록 하는 동작인 경우에는, 자세를 유지하는 상태를 갖도록 하며, 이 경우 각각의 액츄에이터는 기존의 자세를 유지하는 궤적을 따르도록 제어된다. 이러한 경우 로봇에 외력이 주어지면 외력에 따라서 순응형 반응 알고리즘은 적절한 반응 모션을 구현하게 된다. 즉, 앞서 설명한 모션 변화량은 반응 모션으로 나타나게 된다.On the other hand, the first motion by the motion engine 100 is not limited to the one involving movement. In other words, for example, when the first motion by the motion engine 100 is an operation for maintaining a specific posture rather than a movement along a specific trajectory, the posture is maintained. In this case, In accordance with the trajectory. In this case, when an external force is applied to the robot, the adaptive reaction algorithm implements an appropriate response motion according to the external force. That is, the above-described amount of motion variation appears as a reaction motion.
이어서, 외력이 제거되면 기존의 위치로 복귀하여 자세를 유지하게 된다. 예컨대, 로봇이 임의의 자세를 유지하도록 모션 엔진(100)이 제1 모션을 취하는 상태에서 사용자가 로봇의 머리를 쓰다듬는 외력을 가했을 때, 순응형 반응 알고리즘은 이에 순응하는 소정의 모션을 취하게 되며, 외력이 제거되면 다시 원래의 위치로 복귀하여 기존 자세를 유지하는 모션인 제1 모션에 따라서 일정한 자세를 유지하게 된다.Then, when the external force is removed, it returns to the existing position and maintains the posture. For example, when the user applies an external force to strok the head of the robot in a state where the motion engine 100 takes the first motion such that the robot maintains an arbitrary posture, the adaptive reaction algorithm takes a predetermined motion conforming thereto , And when the external force is removed, it returns to the original position and maintains the predetermined posture according to the first motion, which is the motion of maintaining the existing posture.
이러한 예는 도 3 에 도시되어 있다. 먼저, 도 3 의 (a) 와 같이, 로봇의 헤드(T)에 대해 하방향 외력을 가한다고 상정한다. 모션 엔진(100)은 자세를 유지하도록 하는 제1 모션을 생성하고 있으나, 하방향 외력을 감지함에 따라서 순응형 반응 알고리즘은 액츄에이터 궤적을 하방향으로 이동시키는 소정의 순응 모션을 생성한다. 이에 따라서, 로봇의 헤드는 하방향으로 이동하게 된다. An example of this is shown in FIG. First, it is assumed that a downward external force is applied to the head T of the robot as shown in Fig. 3 (a). The motion engine 100 generates the first motion to maintain the attitude, but as the downward external force is sensed, the adaptive reaction algorithm generates a predetermined adaptive motion that moves the actuator locus downward. Accordingly, the head of the robot moves downward.
도 3 의 (b) 와 같이, 외력의 방향이 변하면, 순응형 반응 알고리즘은 변화된 외력에 따라서 제1 모션의 궤적을 변경한다. 아울러, 도 3 의 (c) 와 같이, 외력의 크기가 변하면, 제2 모션 엔진(200)은 외력의 크기의 변화를 감지하여 액츄에이터의 모션 변화량을 작게 하거나 또는 크게 할 수 있다. 이 경우, 제1 모션에 의한 자세 유지 모션이 더 커질 수 있다. As shown in FIG. 3 (b), when the direction of the external force changes, the adaptive reaction algorithm alters the locus of the first motion according to the changed external force. 3 (c), when the magnitude of the external force is changed, the second motion engine 200 may detect a change in the magnitude of the external force and reduce or increase the amount of motion variation of the actuator. In this case, the posture maintaining motion by the first motion can be made larger.
이어서, 도 3 의 (d) 와 같이 외력이 제거되면 제2 모션 엔진(200)에 의한 모션 변화량이 제거되며, 제1 모션에 의해서 초기 위치로 복귀하여 자세를 유지하게 된다. 3 (d), when the external force is removed, the amount of motion variation by the second motion engine 200 is removed, and the first motion returns to the initial position to maintain the posture.
위와 같은 예를 통해 고찰하면, 본 발명에 따른 로봇 제어 시스템(1)은 기본적인 제1 모션을 실시간으로 제어하여 자연스럽게 외력에 순응하는 형태의 모션을 구현할 수 있다.The robot control system 1 according to the present invention can realize a motion in which the first basic motion is controlled in real time to naturally adapt to the external force.
본 발명에 따른 로봇 제어 시스템(1)은, 제1 모션을 생성하는 모션 엔진(100)이 순응형 반응 알고리즘을 가짐으로써, 외력에 따라서 제1 모션을 변형시켜 순응현 모션이 발생하며, 따라서 높은 단계의 로봇과 외부 환경에 대한 상호작용을 구현할 수 있다. The robot control system 1 according to the present invention is a robot control system 1 according to the present invention in which the motion engine 100 generating the first motion has an adaptive reaction algorithm so that the adaptive current motion is generated by modifying the first motion according to the external force, It is possible to implement an interaction with the robot in the external environment.
일 예로, 사용자의 입력 신호 등에 의해서 머리를 끄덕이는 형태의 제1 모션을 수행하는 로봇을 가정한다. 이와 같은 형태의 제1 모션은 모션 엔진(100) 내에 미리 입력된 동작으로서, 사용자의 입력 신호에 의해서 선택된 모션이다. For example, assume a robot that performs a first motion of nodding head by an input signal of a user or the like. The first motion of this type is an operation pre-inputted in the motion engine 100, and is a motion selected by a user's input signal.
머리를 끄덕이는 동작을 수행하기 위해서, 로봇의 각 관절에 구비된 액츄에이터는 시간에 따라서 목표 위치를 따르도록 소정의 궤적을 갖고 제어된다. 예를 들면, 3 개의 관절로 이루어진 목 관절에서, 3 초 후 고개를 30 도 들어 오른 방향 45 도를 바라보는 동작을 하도록 명령이 내려지면 목 관절의 yaw 액츄에이터와 pitch 액츄에이터는 짧은 제어 시간 단위 별로 추종해야 할 모션 값을 생성하여 그에 따른 제어를 수행한다.In order to perform a nodding operation, the actuators provided at the respective joints of the robot are controlled with a predetermined locus so as to follow the target position according to time. For example, in the neck joint consisting of three joints, when the command is given to look at 45 degrees in 30 seconds from the head in three seconds, the yaw actuator and the pitch actuator of the neck joint follow a short control time unit A motion value to be generated is generated, and control according to the generated motion value is performed.
이러한 제1 모션의 수행 도중, 사용자가 로봇의 머리에 외력을 가할 경우를 가정한다. 만일 본 발명과 같은 순응형 반응 알고리즘이 없을 경우, 제1 모션에 대한 변형이 가해지지 않으며, 따라서 로봇은 단순히 고개를 끄덕이는 제1 모션만을 지속하게 된다. 이와 같이 외력에 대한 적절한 대응이 없이 미리 정해진 모션만을 지속적으로 수행하는 로봇의 경우 순응, 회피 등의 동작을 구현할 수 없으며 외력에 대한 강한 반발을 내게 된다. 따라서, 생물체의 동작과 유사한 동작을 구현할 수 없고, 로봇-사람간의 상호작용을 구현할 수 없으며, 로봇-사람간의 정서적인 친밀감과 유대감을 형성시킬 수 없다.During the execution of the first motion, it is assumed that the user applies an external force to the head of the robot. If there is no adaptive reaction algorithm such as the present invention, no deformation is applied to the first motion, and thus the robot simply continues to the first motion nodding the head. In this way, in the case of a robot that continuously performs only predetermined motions without proper response to an external force, operations such as adaptation and avoidance can not be implemented and strong opposition to an external force is caused. Therefore, it is impossible to implement an action similar to that of an organism, to implement a robot-human interaction, and to not create an emotional intimacy and bond between robot and human.
반면에, 본 발명에 의하면, 미리 입력되어 있는 제1 모션을 수행하는 도중 로봇에 외력이 가해지면, 순응형 반응 알고리즘에 제1 모션의 궤적이 실시간으로 변경되게 된다. 따라서, 로봇은 외력에 대한 적절한 상호작용을 할 수 있다.On the other hand, according to the present invention, when an external force is applied to the robot while performing the first motion that is input in advance, the trajectory of the first motion is changed in real time in the adaptive reaction algorithm. Therefore, the robot can make an appropriate interaction with the external force.
또한, 본 발명에 따른 로봇 제어 시스템(1)은 제1 모션 외에 다른 복잡한 모션에 대한 데이터 입력이 없이 단지 제1 모션의 데이터만을 저장하되, 실시간으로 가해지는 외력에 대한 반응은 저장된 순응형 반응 알고리즘에서 제1 모션의 궤적을 변경함으로써 이루어지므로, 즉각적인 반응 모션을 행할 수 있다. 즉, 외력을 받을 경우 다시 명령을 받아 새로운 궤적을 갖는 모션을 따르는 것이 아니라, 기존에 생성되어 있는 제1 모션의 궤적을 실시간으로 변경시키게 되므로, 적은 계산량이 필요하며 순응 모션을 생성하는 데 지연이 발생하지 않게 된다. 또한, 외력이 제거된 경우에는 제1 모션의 궤적으로 복귀하여 재차 통상적인 제1 모션을 수행할 수 있다.In addition, the robot control system 1 according to the present invention stores only the data of the first motion without data input for other complex motions other than the first motion, and the reaction to the external force applied in real- By changing the trajectory of the first motion, it is possible to perform immediate reaction motion. In other words, when receiving an external force, it receives the command again and does not follow the motion having the new trajectory but changes the trajectory of the first motion generated in the real time. Therefore, a small amount of calculation is required, . In addition, when the external force is removed, the motion of the first motion can be restored and the first motion can be performed again.
이상에서는 바람직한 실시예에 대하여 도시하고 설명하였지만, 본 발명은 상술한 특정의 실시예에 한정되지 아니하며, 청구범위에서 청구하는 본 발명의 요지를 벗어남이 없이 당해 발명이 속하는 기술분야에서 통상의 지식을 가진 자에 의해 다양한 변형실시가 가능한 것은 물론이고, 이러한 변형실시들은 본 발명의 기술적 사상이나 전망으로부터 개별적으로 이해되어서는 안될 것이다.While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, but, on the contrary, It should be understood that various modifications may be made by those skilled in the art without departing from the spirit and scope of the invention.

Claims (4)

  1. 하나 이상의 액츄에이터를 포함하는 로봇의 상기 액츄에이터의 작동을 제어하는 로봇 제어 시스템에 있어서,A robot control system for controlling operation of an actuator of a robot including at least one actuator,
    상기 액츄에이터가 소정의 궤적을 갖는 제1 모션을 수행하도록 하는 모션 엔진; 및 A motion engine for causing the actuator to perform a first motion having a predetermined trajectory; And
    상기 로봇에 외력이 가해질 경우, 상기 로봇에 가해진 외력을 센싱하여 상기 외력에 대응하여 상기 제1 모션의 궤적을 변경시키는 순응형 반응 알고리즘을 갖는 모션 변형부;를 포함한 로봇 제어 시스템.And an adaptive reaction algorithm that senses an external force applied to the robot and changes the locus of the first motion corresponding to the external force when an external force is applied to the robot.
  2. 청구항 1에 있어서,The method according to claim 1,
    상기 로봇 제어 시스템은 외력을 센싱하는 센서부; The robot control system includes a sensor unit for sensing an external force;
    상기 센서부에서 센싱된 외력에 대해 소정의 순응 계수 결정 알고리즘에 따라서 순응 계수를 결정하는 순응 계수 결정부;를 더 포함하며,And an adaptive coefficient determiner for determining an adaptive coefficient according to a predetermined adaptive coefficient determination algorithm for an external force sensed by the sensor,
    상기 순응형 반응 알고리즘은, 순응 계수 결정부에서 결정된 순응 계수에 따라서 상기 제1 모션의 궤적을 변경시키는 로봇 제어 시스템Wherein the adaptive reaction algorithm comprises a robotic control system for changing the trajectory of the first motion according to the adaptive coefficient determined by the adaptive coefficient determiner,
  3. 청구항 2에 있어서,The method of claim 2,
    상기 액츄에이터는 소정의 토크값을 갖는 회전 구동기로 구성되며,Wherein the actuator comprises a rotary actuator having a predetermined torque value,
    상기 순응형 반응 알고리즘은,The adaptive reaction algorithm comprises:
    하기 식 1-A 에 따라서 순응 토크를 결정하며,The adaptive torque is determined according to the following equation 1-A,
    TC = TS × C TC = TS x C
    (식 1-A)(Formula 1-A)
    (TC : 순응 토크, TS: 센서부에서 감지된 외력, C : 순응 계수)(TC: adaptive torque, TS: external force sensed by the sensor, C: adaptation coefficient)
    하기 식 2-A 에 따라서 상기 제1 모션의 변화량을 결정하는 로봇 제어 시스템.And determines the amount of change of the first motion according to the following expression (2-A).
    Δθ = (TS-TC)/KS?? = (TS-TC) / KS
    (식 2-A) (Formula 2-A)
    (Δθ : 제1 모션의 변화량, KS : 시스템의 강성 계수)(??: change amount of the first motion, KS: stiffness coefficient of the system)
  4. 청구항 2에 있어서,The method of claim 2,
    상기 액츄에이터는 소정의 직선 구동기로 구성되며,The actuator is constituted by a predetermined linear actuator,
    상기 순응형 반응 알고리즘은,The adaptive reaction algorithm comprises:
    하기 식 1-B 에 따라서 순응 힘을 결정하며,Determines the compliance force in accordance with Equation 1-B below,
    FC = FS × C FC = FS × C
    (식 1-B)(Formula 1-B)
    (FC : 순응 힘, FS: 센서부에서 감지된 외력, C : 순응 계수)(FC: conforming force, FS: external force detected by the sensor section, C: adaptation coefficient)
    하기 식 2-B 에 따라서 상기 제1 모션의 변화량을 결정하는 로봇 제어 시스템.Wherein the amount of change of the first motion is determined according to the following equation (2-B).
    ΔSP = (FS-FC)/KA? SP = (FS-FC) / KA
    (식 2-B) (Formula 2-B)
    (ΔSP : 제1 모션의 변화량, KA : 시스템의 강성 계수)(? SP: change amount of the first motion, KA: stiffness coefficient of the system)
PCT/KR2017/010883 2017-08-25 2017-09-28 Robot control system WO2019039646A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
KR10-2017-0108141 2017-08-25
KR1020170108141A KR102001101B1 (en) 2017-08-25 2017-08-25 Robot controlling system

Publications (1)

Publication Number Publication Date
WO2019039646A1 true WO2019039646A1 (en) 2019-02-28

Family

ID=65440077

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/KR2017/010883 WO2019039646A1 (en) 2017-08-25 2017-09-28 Robot control system

Country Status (2)

Country Link
KR (1) KR102001101B1 (en)
WO (1) WO2019039646A1 (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006000955A (en) * 2004-06-16 2006-01-05 National Institute Of Advanced Industrial & Technology Robot arm, and its rotating joint device and wrist device
JP2009018380A (en) * 2007-07-12 2009-01-29 Toyota Motor Corp Robot, control method of the robot, and control system of the robot
KR101227092B1 (en) * 2010-11-05 2013-01-29 한국과학기술연구원 Motion Control System and Method for Robot
KR101321791B1 (en) * 2012-01-02 2013-10-28 한국과학기술연구원 Robot Motion Control System and Method for Active Body Rehabilitation
JP2016153155A (en) * 2015-02-20 2016-08-25 ファナック株式会社 Human-robot cooperative system causing robot to evacuate

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR100880613B1 (en) 2007-08-10 2009-01-30 주식회사 쿼터뷰 System and method for supporting emotional expression of intelligent robot and intelligent robot system using the same
KR101348940B1 (en) * 2012-06-26 2014-01-09 한국과학기술연구원 Sensory Reproduction System to transmit physical intercations in whole-body remote control of Robot
US9226796B2 (en) * 2012-08-03 2016-01-05 Stryker Corporation Method for detecting a disturbance as an energy applicator of a surgical instrument traverses a cutting path

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006000955A (en) * 2004-06-16 2006-01-05 National Institute Of Advanced Industrial & Technology Robot arm, and its rotating joint device and wrist device
JP2009018380A (en) * 2007-07-12 2009-01-29 Toyota Motor Corp Robot, control method of the robot, and control system of the robot
KR101227092B1 (en) * 2010-11-05 2013-01-29 한국과학기술연구원 Motion Control System and Method for Robot
KR101321791B1 (en) * 2012-01-02 2013-10-28 한국과학기술연구원 Robot Motion Control System and Method for Active Body Rehabilitation
JP2016153155A (en) * 2015-02-20 2016-08-25 ファナック株式会社 Human-robot cooperative system causing robot to evacuate

Also Published As

Publication number Publication date
KR102001101B1 (en) 2019-07-18
KR20190022174A (en) 2019-03-06

Similar Documents

Publication Publication Date Title
Li et al. Adaptive human–robot interaction control for robots driven by series elastic actuators
Okamura et al. An overview of dexterous manipulation
KR101549818B1 (en) Robot hand and method of controlling robot hand
US10220512B2 (en) Position/force controller, and position/force control method and program
Laghi et al. Shared-autonomy control for intuitive bimanual tele-manipulation
Li et al. Survey on mapping human hand motion to robotic hands for teleoperation
Platt et al. Multiple-priority impedance control
Onishi et al. Generation of human care behaviors by human-interactive robot RI-MAN
Sun et al. Single master bimanual teleoperation system with efficient regulation
Zollo et al. An experimental study on compliance control for a redundant personal robot arm
JP2010142909A (en) Controller for robot
JP2010069584A (en) Device and method for controlling manipulator
Si et al. Adaptive compliant skill learning for contact-rich manipulation with human in the loop
Santiago et al. Human-inspired stable bilateral teleoperation of mobile manipulators
Su et al. Sensor fusion-based anthropomorphic control of under-actuated bionic hand in dynamic environment
Koropouli et al. Learning interaction control policies by demonstration
WO2019039646A1 (en) Robot control system
Pistillo et al. Bilateral physical interaction with a robot manipulator through a weighted combination of flow fields
Gao et al. Design and analysis of underactuated robotic gripper with adaptive fingers for objects grasping tasks
Mittendorfer et al. Self-organizing sensory-motor map for low-level touch reactions
Frigola et al. Human-robot interaction based on a sensitive bumper skin
Pairet et al. Learning and composing primitive skills for dual-arm manipulation
Sakaino Bilateral control-based imitation learning for velocity-controlled robot
Papageorgiou et al. Task geometry aware assistance for kinesthetic teaching of redundant robots
Buss et al. Advanced telerobotics: Dual-handed and mobile remote manipulation

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17922394

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 17922394

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205N DATED 24/06/2020)

122 Ep: pct application non-entry in european phase

Ref document number: 17922394

Country of ref document: EP

Kind code of ref document: A1