Neural Network Force Control for Industrial Robots

In this paper, we present a hierarchical force control framework consisting of a high level control system based on neural network and the existing motion control system of a manipulator in the low level. Inputs of the neural network are the contact force error and estimated stiffness of the contacted environment. The output of the neural network is the position command for the position controller of industrial robots. A MITSUBISHI MELFA RV-M1 industrial robot equipped with a BL Force/Torque sensor is utilized for implementing the hierarchical neural network force control system. Successful experiments for various contact motions are carried out. Additionally, the proposed neural network force controller together with the master/slave control method are used in dual-industrial robot systems. Successful experiments are carried out for the dual-robot system handling an object.     

Decentralized adaptive coordinated control of multiple robot arms without using a force sensor

This paper presents a distributed adaptive coordinated control method for multiple robot arms grasping a common object. The cases of rigid contact and rolling contact are analyzed. In the proposed controller, the dynamic parameters of both object and robot arms are estimated adaptively. The desired motions of the robot arms are generated by an estimated object reference model. The control method requires only the measurements of the positions and velocities of the object and robot arms, but not the measurements of forces and moments at contact points. The asymptotic convergence of trajectory is proven by the Lyapunov-like Lemma. Experiments involving two robot arms handling a common object are shown.     

自适应模糊与CMAC并行的机器人力/位置控制

为提高机器人系统对机器人末端操纵器与外界工作环境接触时，其接触刚度不确定性的自适应能力，在机器人力／位置混合控制的基础上，设计出了一种基于自适应模糊与CMAC并行控制的机器人力控制器，采用小脑模型神经控制器实现前馈控制，实现被控对象的逆动态模型，自适应模糊控制器实现反馈控制，保证系统的稳定性，且抑制扰动。以平面两关节机器人进行仿真，仿真结果表明，系统的自适应能力和力跟踪能力有显著的提高，机械手在其末端操纵器与刚性变化范围较大的外界工作环境接触时，具有较强的适应能力，较好地完成了机器人的力／位置控制。     

机器人触须传感器动态工作机理的研究

提出一种测量机器人触须与被测物体接触位置的新方法。在弹性触须的顶端加上适当的重物,增加触须的摆动惯性。当触须与物体发生接触后,触须停止摆动,触须的运动状态在与物体接触、非接触状态之间返复变化。信号采集系统以图形方式实时输出触须的振动波形,通过测量触须的振动频率,即可得出触须与物体的接触位置。     

Adaptive hybrid control of manipulators on uncertain flexible objects

This paper presents an adaptive hybrid control approach for a robot manipulator to interact with its flexible object. Because of its flexibility, the object dynamics influence the robot's control system, and since it is usually a distributed parameter system, the object dynamics as seen from the robot change when the robot moves. The problem becomes further complicated such that it is difficult to decompose the robot's position and contact force control loops. In this paper, we approximate the object's distributed parameter model into a lumped 'position state-varying' model. Then, by using the well-known nonlinear feedback compensation, we decompose the robot's control space into a position control subspace and object torque control subspace. We design the optimal state feedback for the position control loop and control the robot's contact force through controlling the resultant torque of the object. We use the model-reference simple adaptive control strategy to control the torque control loop. We also study the problem on how to select a reasonable reference model for this control loop. Experiments of a PUMA robot interacting with an aluminum beam show the effectiveness of our approach.     

Robot link and contact surface dynamic interaction: An experimental study of contact task instability

In this article we explore, both theoretically and experimentally, the effect of multiple structural compliance sources on robot stability during contact task execution. Here we examine the effect of link compliance and compliance of the contact surface on overall system stability. Using the theory associated with singularly perturbed dynamic systems, the stability of the robotic system is examined for three distinct cases: (1) link stiffness much greater than contact surface stiffness; (2) contact surface stiffness much greater than link stiffness; and (3) link stiffness comparable to surface stiffness. Theoretical results lead to a conclusion of stability only for case (1). Experiments indicate stable behavior for all these cases. Through the use of singular perturbation modeling and analysis, we are able to write dynamic models of the robot during contact with a complaint surface such that the complaint behavior of the robot links and contact surface are parameterized with just two variables. In doing so, theoretical examination and experimental study of stability, or the lack thereof, is greatly simplified. Instead of exhaustive tests in which experiments or stability analyses are conducted over a wide range of values of link and contact surface compliance, a select number of cases are examined. Under the assumptions that validate the use of singular perturbation modeling techniques, the resultant stability analysis and experimental results are hence conducted in an efficient manner. © 1997 John Wiley & Sons, Inc.     

Estimation of Joint Stiffness Using Instantaneous Loads via an Electromyogram and Application to a Master–Slave System with an Artificial Muscle Manipulator

Master–slave systems and human-assisted systems in which robots can work together with humans have been widely developed. Such systems are necessary to communicate human intentions to a robot. Therefore, it is important for these systems to be able to estimate human joint characteristics such as torque, position and stiffness. In this context, we focus on the myoelectric (ME) potential. In many previous studies, an electric motor has been used as an actuator, and the torque and position are estimated from the ME potential. However, the joint stiffness of humans has not been studied extensively. In this study, we propose a method to estimate the stiffness of the human elbow by using antagonistic muscles when an instantaneous load is applied. Furthermore, we apply our method to a 1-d.o.f. manipulator with an artificial muscle. In the case of eccentric contraction, it has been shown that the stiffness of a joint can be estimated solely by an electromyogram of the triceps. Then, the stiffness and angle control of the artificial muscle manipulator that used it as the slave side is proposed. Furthermore, the estimated joint stiffness is set as a desired value for joint stiffness control of the artificial muscle manipulator. Experimental results of stiffness control indicate that the angle and the stiffness of the 1-d.o.f. artificial muscle manipulator can be adequately controlled for a master–slave system. Safer remote control systems can be developed by using this system.     

Normal contact stiffness identification-based force compensation for a hardware-in-the-loop docking simulator

Docking is a very important and challenging space operation. To ensure the safety and reliability of the docking, the thorough ground verification is necessary. Hardware-in-the-loop (HIL) simulation is a good choice, especially when the geometric structure of the docking device is complex. However, the HIL simulation suffers from energy increase and instability since time delay in the system is unavoidable. In this study, a six degree-of-freedom (DOF) parallel robot is used as the motion simulator to physically perform the contact in the simulation. To achieve system stability and high fidelity, a normal contact stiffness identification-based force compensation method is proposed. The contact force is compensated through identifying the time-varying stiffness along the normal direction of the contact surface. The 6-DOF force-moment compensation algorithm for the HIL parallel robot system is presented in detail. Simulations and experiments are both carried out, and the results show that the proposed compensation method is effective.     

Measurement-based modeling of contact forces and textures for haptic rendering

Haptic texture represents the fine-grained attributes of an object's surface and is related to physical characteristics such as roughness and stiffness. We introduce an interactive and mobile scanning system for the acquisition and synthesis of haptic textures that consists of a visually tracked handheld touch probe. The most novel aspect of our work is an estimation method for the contact stiffness of an object based solely on the acceleration and forces measured during stroking of its surface with the handheld probe. We establish an experimental relationship between the estimated stiffness and the contact stiffness observed during compression. We also measure the height-displacement profile of an object's surface enabling us to generate haptic textures. We show an example of mapping the textures on to a coarse surface mesh obtained with an image-based technique, but the textures may also be combined with coarse surface meshes obtained by manual modeling.     

Fast grasping of unknown objects through automatic determination of the required number of mobile robots

In this paper, we present a strategy for fast grasping of unknown objects by mobile robots through automatic determination of the number of robots. An object handling system consisting of a Gripper robot and a Lifter robot is designed. The Gripper robot moves around an unknown object to acquire partial shape information for determination of grasping points. The object is transported if it can be lifted by the Gripper robot. Otherwise, if all grasping trials fail, a Lifter robot is used. In order to maximize use of the Gripper robot’s payload, the detected grasping points that apply the largest force to the gripper are selected for the Gripper robot when the object is grasped by two mobile robots. The object is measured using odometry and scanned data acquired while the Gripper robot moves around the object. Then, the contact point for calculating the insert position for the Lifter robot can be acquired quickly. Finally, a strategy for fast grasping of known objects by considering the transition between stable states is used to realize grasping of unknown objects. The proposed approach is tested in experiments, which find that a wide variety of objects can be grasped quickly with one or two mobile robots.     

Interaction model between elastic objects for haptic feedback considering collisions of soft tissue

The simulation of organ–organ interaction is indispensable for practical and advanced medical VR simulator such as open surgery and indirect palpation. This paper describes a method to represent real-time interaction between elastic objects for accurate force feedback in medical VR simulation. The proposed model defines boundary deformation of colliding elements based on temporary surface forces calculated by temporary deformation. The model produces accurate deformation and force feedback considering collisions of objects as well as prevents unrealistic overlap of objects. A prototype simulator of rectal palpation is constructed on general desktop PC with a haptic device, PHANToM. The system allows users to feel different stiffness of a rear elastic object located behind another elastic object. The results of experiments confirmed the method expresses organ–organ interaction in real-time and produces realistic and perceivable force feedback.     

Development of an Underwater Robotic Inspection System using Mechanical Contact

This paper reports the development of a robotic inspection system using a mechanical contact mechanism that enhances the positioning stability of a small and lightweight underwater robot to take clear images of underwater targets and to work with manipulators for inspections under external disturbances. As described in this paper, first we perform a two‐dimensional numerical analysis based on force and moment acting on an underwater robot with a contact mechanism. Second, we experimentally investigate the friction coefficients of several soft and high friction materials for the contact points of a prototype contact mechanism to enhance the positioning stability of the robot. Based on the results of numerical analysis and the experimental investigation, we design and develop a prototype contact mechanism for an underwater robot. Moreover, we experimentally test the stability of the underwater robot with the contact mechanism in a test tank. Finally, a ship hull inspection is conducted as a field test in a port using the robot with the developed contact mechanism. The experimentally obtained results indicate that the proposed contact mechanism is a useful tool for underwater visual inspections and manipulator tasks of a small and lightweight underwater robot.     

A Generalized Vision-based Stiffness Controller for Robot Manipulators with Bounded Inputs

Generally, stiffness and impedance control schemes require knowledge of the location of any object with which a robot interacts within its workspace; therefore, the integration of a computer vision system within the control loop allows us to know the location of the robot end effector and the object (target) simultaneously. In this paper, a generalized and saturating vision-based stiffness controller with adaptive gravity compensation is presented. The proposed control algorithm is designed to regulate robot-environment interaction in task-space, where the contact force is modeled as a vector of generalized bounded spring-like forces. In order to control nonredundant robots, the proposed controller has a nonlinear proportional-derivative structure with static model-based compensation of gravitational forces, as it includes a regressor-based adaptive term. To support the proposal, the Lyapunov stability analysis of the closed-loop equilibrium vector is presented. Finally, the suitable performance of the proposed scheme was verified by numerical simulations and experimental tests.

     

A force reflected exoskeleton-type masterarm for human-robot interaction

Two human-robot interactions, including a haptic interaction and a teleoperated interaction, are explored with a new exoskeleton-type masterarm, in which the electric brakes with the torque sensor beams are used for force reflection. In the haptic interaction with virtual environment, the masterarm is used as a haptic device and tested to examine how the resistant torque of the electric brake for the force reflection is implemented in contact regime prior to conducting the teleoperated interaction. Two types of virtual environments, a rigid wall with high stiffness (hard contact with 10 [KN/m]) and a soft wall with low stiffness (soft contact with 0.1 [N/m]), are integrated with the masterarm for the haptic interaction. In hard contact, large force is fed back to the human operator, and makes the human operator hardly move. The electric brake with the torque sensor beam can detect the torque and its direction so that it allows free motion as well as contact motion by releasing or holding the movement of the operator. The experimental results show how the electric brake is switched from contact to free regime to allow the operator to move freely, especially when the operator intends to move toward the free regime in contact. In soft contact, the force applied to the human operator can be increased or decreased proportionally to the torque amount sensed by the torque sensor beam, thus the operator can feel the contact force proportional to the amount of the deformation during the contact. Finally, the masterarm is integrated with the humanoid robot, CENTAUR developed at Korea Institute of Science and Technology to conduct a pick-and-place task through the teleoperated interaction. It is examined that the CENTAUR as a slave robot can follow the movement of the operator.     

Localization of a mobile robot based on an ultrasonic sensor using dynamic obstacles

In this article, we propose a localization scheme for a mobile robot based on the distance between the robot and moving objects. This method combines the distance data obtained from ultrasonic sensors in a mobile robot, and estimates the location of the mobile robot and the moving object. The movement of the object is detected by a combination of data and the object’s estimated position. Then, the mobile robot’s location is derived from the a priori known initial state. We use kinematic modeling that represents the movement of a robot and an object. A Kalman-filtering algorithm is used for addressing estimation error and measurement noise. Throughout the computer simulation experiments, the performance is verified. Finally, the results of experiments are presented and discussed. The proposed approach allows a mobile robot to seek its own position in a weakly structured environment. This work was presented in part at the 12th International Symposium on Artificial Life and Robotics, Oita, Japan, January 25–27, 2007     

Robust bin-picking system using tactile sensor

ABSTRACT

This paper presents a robust bin-picking system utilizing tactile sensors and a vision sensor. The object position and orientation are estimated using a fast template-matching method through the vision sensor. When a robot picks up an object, the tactile sensors detect the success or failure of the grasping, and a force sensor detects the contact with the environment. A weight sensor is also used to judge whether the lifting of the object has been successful. The robust and efficient bin-picking system presented herein is implemented through the integration of different sensors. In particular, the tactile sensors realize rope-shaped object picking that has yet to be made possible with conventional picking systems. The effectiveness of the proposed method was confirmed through grasping experiments and in a competitive event at the World Robot Challenge 2018.     

A rod-mass method of soft tissue modeling and three dimensional force vector estimation for tele-surgery training

In a tele-surgery training system, the transparency is extremely important so as to ensure the success of the operation and the safety of soft objects. Due to current technique limits, it is difficult to mount force sensors at the end of the slave manipulator. In this paper, we propose a novel rod-mass algorithm and construct the model of soft objects. Through the modeling process, the accurate three dimensional contact force vector between the end of the manipulator and the soft object can be estimated in real time. A virtual spring using Hooke s law is introduced to the novel mass–spring method. Applying an impedance model, the three dimensional contact force estimates can be calculated from the deformation of masses’ positions and velocities. In order to verify our methods, a virtual reality interaction platform is constructed including the Omni master manipulator, a four joints manipulators, a virtual reality display, and the soft object’s model. Numerical simulations and experiments are performed to verify the accuracy and the feasibility of soft objects grasping. Results show the high effectiveness and efficiencies of our methods.     

Dynamic grasp planning of multifingered robot hands based onasymptotic stability

We describe an approach for planning grasps of multifingered robot hands based on a small vibration model. Using features of the grasp configuration, we analyze asymptotic stability, contact situations, and uniaxial fingertip force constraints for the combined planning of finger posture and finger position, and characterize the generalized mass, damping, and stiffness. Choosing the largest time constant of the vibration model as an optimization criterion for planning finger postures and positions, the original problem of dynamic grasp planning is formulated as a nonlinear program. Simulation examples for a three-fingered robot hand grasping a spherical object demonstrate the effectiveness of the approach.     

Object handling control among two-wheel robots and a human operator: An empirical approach

This article presents object handling control between two-wheel robot manipulators, and a two-wheel robot and a human operator. The two-wheel robot has been built for serving humans in the indoor environment. It has two wheels to maintain balance and is able to make contact with a human operator via an object. A position-based impedance force control method is applied to maintain stable object-handling tasks. As the human operator pushes and pulls the object, the robot also reacts to maintain contact with the object by pulling and pushing against the object to regulate a specified force. Master and slave configuration of two-wheel robots is formed for handling an object, where the master robot or a human leads the slave robot equipped with a force sensor. Switching control from position to force or vice versa is presented. Experimental studies are performed to evaluate the feasibility of the object-handling task between two-wheel mobile robots, and the robot and a human operator.     

Position estimation of a mobile robot using images of a moving target in intelligent space with distributed sensors

Latest advances in hardware technology and state-of-the-art of mobile robots and artificial intelligence research can be employed to develop autonomous and distributed monitoring systems. A mobile service robot requires the perception of its present position to co-exist with humans and support humans effectively in populated environments. To realize this, a robot needs to keep track of relevant changes in the environment. This paper proposes localization of a mobile robot using images recognized by distributed intelligent networked devices in intelligent space (ISpace) in order to achieve these goals. This scheme combines data from the observed position, using dead-reckoning sensors, and the estimated position, using images of moving objects, such as a walking human captured by a camera system, to determine the location of a mobile robot. The moving object is assumed to be a point-object and projected onto an image plane to form a geometrical constraint equation that provides position data of the object based on the kinematics of the ISpace. Using the a priori known path of a moving object and a perspective camera model, the geometric constraint equations that represent the relation between image frame coordinates for a moving object and the estimated robot's position are derived. The proposed method utilizes the error between the observed and estimated image coordinates to localize the mobile robot, and the Kalman filtering scheme is used for the estimation of the mobile robot location. The proposed approach is applied for a mobile robot in ISpace to show the reduction of uncertainty in determining the location of a mobile robot, and its performance is verified by computer simulation and experiment.