Advances in Robotic Peg-in-Hole Assembly: A Comprehensive Review

This paper proposes a simple algorithm for generating a 3-D path of an end effector of a robot arm in real-time. We assume that the global position and orientation of the robot's end effector, a waypoint and a destination are known. Our algorithm computes a 2-D dipole field on the local plane through the end effector, the waypoint and the destination. The robot's end effector moves along the 2-D dipole field. Through experiments in computer simulation and real environment, we confirm the validity of our method.

Currently used robot position control systems consider only the joint angles when positioning an end effector in space. This approach can lead to serious errors due to structural deflection under load, backlash in joints, manufacturing tolerances, etc. A method is needed to directly measure the absolute position and orientation of the end effector in three-dimensional space in order to achieve positioning accuracy. An ultrasonic ranging system is proposed to accomplish this task. It consists of: (1) one or more transmitters mounted on the robot's end effector; (2) receivers (four or more required), mounted adjacent to the robot's work envelope; and (3) analog and digital electronics to drive the transducers, process the received signals, and calculate the position of the end effector. The end effector's position is calculated using the time of flight measured from the emitter to each receiver, by an intersection of spheres algorithm. The performance objectives of this system are: (a) measure position with accuracy of ± 0.001 in. (b) up to a range of 3 m. and (c) with at least 1-Hz update rate. A theoretical error analysis of the system is presented. Types of emitter signals are compared, operating frequency considerations are discussed, and hardware recommendations are made. [Work supported by NSF.]

Deep visual regression models have an important role to find how much the learning model fits the relationship between the visual data (images) and the predicted continuous output. Recently, deep visual regression has been utilized in different applications such as age prediction, digital holography, and head-pose estimation. Deep learning has recently been cutting-edge research. Most of the research papers have focused on utilizing deep learning in classification tasks. There is still a lack of research that use deep learning for regression. This paper utilizes different deep learning models for two regression tasks. The first one is the prediction of the image rotation angle. The second task is to predict the position of the robot's end-effector in 2D space. Efficient features were learned or extracted in order to perform good regression. The paper demonstrates and compares various models such as a local Receptive Field-Extreme Learning Machine (LRF-ELM), Hierarchical ELM, Supervised Convolutional Neural Network (CNN), and pre-trained CNN such as AlexNet. Each model was trained to learn or extract features and map them to specific continuous output. The results show that all models gave good performance in terms of RMSE and accuracy. H-ELM was found to outperform other models in term of training speed.

Robot end-effector position measurement is very essential to determine the error between the commanded position and actual position reached by the robot end-effector. This also ensures the error with which the robot end-effector traces the commanded path. Conventionally position measuring devices like laser trackers, coordinate measuring machine and other non-contact-based position measuring devices are widely used to measure the position of the end-effector due to their high accuracy and precision, even though they are very expensive. This paper presents a cost effective solution for the measurement of position of robot end-effector. An end-effector position measuring device is designed and developed in spherical coordinate system using two absolute rotary encoders and a draw-wire sensor. A mathematical model for the end effector position measuring device is developed to determine the position of the end-effector with respect to a reference coordinate system using 3D homogeneous transformation approach. End-effector position measuring device is tested for ABB IRB 1600 for numerous poses and for a straight line path. The newly developed end-effector position measuring device is found to be capable of measuring the end effector position with an accuracy of 2 mm and hence can be implemented in calibration of industrial robots.

Robot end-effector position measurement is very essential to determine the error between the commanded position and actual position reached by the robot end-effector. This also ensures the error with which the robot end-effector traces the commanded path. Conventionally position measuring devices like laser trackers, coordinate measuring machine and other non-contact-based position measuring devices are widely used to measure the position of the end-effector due to their high accuracy and precision, even though they are very expensive. This paper presents a cost effective solution for the measurement of position of robot end-effector. An end-effector position measuring device is designed and developed in spherical coordinate system using two absolute rotary encoders and a draw-wire sensor. A mathematical model for the end effector position measuring device is developed to determine the position of the end-effector with respect to a reference coordinate system using 3D homogeneous transformation approach. End-effector position measuring device is tested for ABB IRB 1600 for numerous poses and for a straight line path. The newly developed end-effector position measuring device is found to be capable of measuring the end effector position with an accuracy of 2 mm and hence can be implemented in calibration of industrial robots.

This article discusses the innovative application of a Cartesian robot manipulator with acoustic feedback for calibration and precise positioning of a string-excitation element in investigating stringed instruments. It describes an experiment in which an acoustic guitar string is automatically excited with different guitar picks. The robot's end effector positioning system utilizes limit switches, acting as a mechanical sensor, which provides feedback to the linear actuators that are equipped with stepper motors. The authors detail the research challenges they faced and propose a positioning algorithm that makes use of a microphone as an acoustic sensor, improving the calibration of the end effector's position.

Graspless manipulation is easily interfered by external disturbances because the manipulated object is not completely held by a robot hand and supported by an environment such as a floor. Thus it is important to ensure the manipulation is executed robustly against some disturbances. In our works, a rigid-body-based analysis of indeterminate contact forces for quasi-static graspless manipulation has been proposed, and also joint torque optimization for robotic hands. The joint torques of the robot is determined in consideration of some robustness of manipulation against disturbances, which include changes or estimation errors of friction. In the analysis of contact forces in quasi-statics, a kinematic constraint on static friction is considered to exclude infeasible sets of frictional force, with considering treatment of kinetic friction. Additionally, new objective functions for computing optimal joint torques in both static and quasi-static graspless manipulation are proposed. Some numerical samples of both applications are shown to verify our proposed methods.

Toward the goal of robots performing robust and intelligent physical interactions with people, it is crucial that robots are able to accurately sense the human body, follow trajectories around the body, and track human motion. This study introduces a capacitive servoing control scheme that allows a robot to sense and navigate around human limbs during close physical interactions. Capacitive servoing leverages temporal measurements from a multielectrode capacitive sensor array mounted on a robot's end effector to estimate the relative position and orientation (pose) of a nearby human limb. Capacitive servoing then uses these human pose estimates from a data-driven pose estimator within a feedback control loop in order to maneuver the robot's end effector around the surface of a human limb. We provide a design overview of capacitive sensors for human–robot interaction and then investigate the performance and generalization of capacitive servoing through an experiment with 12 human participants. The results indicate that multidimensional capacitive servoing enables a robot's end effector to move proximally or distally along human limbs while adapting to human pose. Using a cross-validation experiment, results further show that capacitive servoing generalizes well across people with different body size.

The fulfillment of programmed tasks by industrial robots in redundant conditions occurs when a robotic system has more degrees of freedom than would be necessary to accomplish the task. From the point of view of mechanical modeling of robotic structures, this translates into the fact that the number of motor coordinates of the robot is greater than would be necessary and sufficient to ensure the position and orientation of the robot's end effector required by the task. In such situations, there is the possibility of optimally performing the programmed task under the conditions of meeting certain criteria that may refer to better accessibility of the robot's workspace in the presence of obstacles, achieving a minimum in terms of movements at the level of the robot's motor axes or obtaining motor coordinate values as far as possible from the edges of their variation ranges. The paper analyzes the case of a robot with five rotation modules whose task is to ensure a trajectory imposed at the origin of the coordinate system attached to its end effector. The redundancy case is analyzed by imposing as an optimization criterion the achievement of a minimum of displacements at the motor axes level of the analyzed robot. Finally, the results obtained for the variation of motor coordinates in the case of a linear interpolation of the end effector movement are presented.

This article describes a laser tracking system (LTS) that can be used to determine the position and orientation of a robot's end effector with high accuracy during arbitrary robot motions. The position is measured using a polar configuration employing one laser beam and two rotary axes. A retroreflector in the robot end effector reflects the laser beam and constitutes the only part the robot has to carry for the contactless measurement. The orientation is determined by analyzing the intensity profile of the reflected laser beam with a vision system. The intensity profile carries diffraction patterns of the retroreflector edges that uniquely define orientation. The tracking unit allows the system to follow arbitrary movements of the robot. With these characteristics a six-degree-of-freedom (6-DOF) real-time robot measurement system is provided that can dynamically track robot motions.

Chapter Two - Inverse kinematics analysis of 7-degree of freedom welding and drilling robot using artificial intelligence techniques

In this paper, an adaptive hybrid force/position control architecture for a cooperative system consisting of two unknown planar manipulators grasping and manipulating an unknown rigid object is proposed. It is assumed that the object will be in contact with an environment through an assembly process. The contact force between each robot and the object is decomposed into two independent forces, one of which is for grasping the object, and the other is for motion of the object. The reduced order dynamics of the cooperative system is obtained with respect to the position of the first robot's end effector. The object's position and interaction with the environment is controlled using an adaptive hybrid control algorithm. The Lyapunov's direct method is used in design and stability analysis of the proposed controller. Simulation results, confirm the performance and effectiveness of the proposed methodology.

Presents a new approach to online trajectory planning for target tracking, dynamic grasping and catching. The robot executes a geometric controller-it simply evaluates a nonlinear function which maps the currently measured position and velocity of the object to be grasped into a current desired robot pose. If the robot tracks these setpoints, it is guaranteed to match the object's velocity and acceleration on a specified grasp surface. The authors develop a geometric controller which specifies the full 6DOF position and orientation of the robot's end effector. A planar simulation demonstrates that this paradigm performs favorably when compared with the traditional planning approach. Since it does not depend on future object measurements, no object model is needed for trajectory prediction. Without trajectory prediction, the computational effort is drastically reduced, allowing for higher controller speed and tracking feedback gains. At the same time this approach provides a framework for general sensor based control of robotic tasks. >

In this study, a gantry type welding robot having three prismatic and two rotational joints was used. By creating the kinematic diagram of this robot in Cartesian space its inverse kinematic equations were obtained. Denavit Hartenberg rules defining the movement of one-link relative to another, were applied in drawing the kinematic diagram. The D-H method provides great easiness in forward and inverse kinematics calculations. With this method, the D-H parameters table to be used in kinematic calculations was created and inverse kinematic equations were obtained. Using inverse kinematic equations, the known position and orientation of the robot's end effector and the parameters of the position and orientation of each link were obtained. All these kinematic calculations were performed with a user interface software (GUI) prepared in Microsoft Visual Studio C# program. In this software, Mach3 program was also used as an assistant to control the motors with the position and orientation information obtained for each motor. In this way, a smooth welding application in the desired position and orientation is aimed.

Employment of robots in manufacturing has been a value-added entity in a manufacturing industry. Robotic simulation is used to visualize entire robotic application system, to simulate the movement of robot arm incorporated with components consist in its environment and to detect collision between the robot and components. This paper presents result of a project in implementing a computer based model to simulate Okura A1600 palletizer robot. The application uses Okura A1600 robot for palletizing bags at the end of the production line and focuses on pick-and-place application. The project objective is to generate a computer simulated model to represent the actual robot model and its environment. The project simulates the robot's first four joints, namely as the waist, shoulder, elbow and waist and focuses on the position of the robot's end effector, regardless its orientation. Development of the model is using Workspace5 as a simulation tool. Two types of methodology are used, which are the methodology for developing the robotic workcell simulation model and the methodology for executing the robotic simulation. The output of the project will be a three-dimensional view of robot arm movement based on series of predefined geometry points, layout checking and robot's reachability by generating working envelope, collision and near miss detection, and monitoring on the cycle time upon completing a task. The project is an offline programming and no robot language is generated.