Tadeusz Szkodny

Gesture based robot control

The paper proposes a method of controlling robotic manipulators with use of human gestures and movement. Experiments were performed with the use of 4 degree-of-freedom AX-12 Robotic Arm manipulator with force gripper and ASUS Xtion depth sensor also called motion controller. Depth and video capture has been done via OpenNI library. The infrastructure is based on Windows Communication Foundation (WCF) for remote access, authorization, multimedia streaming and servo control. Control of robotic manipulator is implemented with use of human computer interaction algorithm basing on depth sensor information.

Basic Component of Computational Intelligence for IRB-1400 Robots

In this paper the solution algorithm of the inverse kinematics problem for IRB-1400 robots is presented. This algorithm may be a basic component of future computational intelligence for theses robots. The problem of computing the joint variables corresponding to a specified location of the end-effector is called the inverse kinematics problem. This algorithm may be used with the current computer control algorithms for these robots. It would allow controlling these robots through the use of the vision system, which specifies the required location of the end-effector. This required location makes it possible for the end-effector to approach an manipulation object (observed by the vision system) and pick it up. These robots are equipped with several manipulator which has 6 links connected by the revolute joints. First the location of the end-effector in relation to the base of the manipulator is described. Next the analytical formulas for joint variables (dependent on these location) are derived. These formulas have take into account the multiple solutions for the singular configurations of this manipulator.

Planning and Realization of Singular Trajectories of Adept Six 300 Robot

In this paper the solutions of forward and inverse kinematics problem for Adept Six 300 robot are presented. A matrix method was used in computations. Presented algorithm was used in trajectory planning. Joint values are chosen from a given set of solutions. A graph of solutions is made. Nodes represent the solutions of inverse kinematics problem and arcs’ weight are the cost of transition between two configurations. Two criteria are proposed in order to obtain smooth motion with desired properties. Algorithm based on derived formulas and logical conditions has been implemented in Adept SmartController CX. A comparison between manufacturer’s and proposed algorithm is presented. Proposed solution of inverse kinematics problem allows for a smooth passage through singular points.

Avoiding of the Kinematic Singularities of Contemporary Industrial Robots

The singular configurations of contemporary industrial robot manipulators of such renowned companies as ABB, Fanuc, Mitsubishi, Adept, Kawasaki, COMAU and KUKA, cause undesired stopping.It is the basic defect of these robots. The paper presents simple method of avoiding these singularities. To determine the singular configurations of these manipulators a global form of description of the end-effector kinematics was prepared, relative to the other links.On the base of this description , the formula for the Jacobian was defined, in the end-effector coordinates. Next, a closed form of the determinant of Jacobian was derived.From the formula, singular configurations, where the determinant’s value equals zero, were determined.Additionally, geometric interpretations of these configurations were given and they were illustrated. For the exemplary manipulator, small corrections of joint variables preventing the reduction of the Jacobian order were suggested. An analysis of positional errors, caused by these corrections,was presented.

Remote Control and Monitoring of AX-12 Robotic Arm Based on Windows Communication Foundation

The paper proposes a service-oriented architecture system for control and state monitoring of robotic manipulators. Experiments were performed with the use of 4 degree-of-freedom AX-12 Robotic Arm manipulator with gripper and laser effector, as well as a high resolution GoPro Hero HD camera and frame grabber. Multimedia device management and video capture has been done via DirectShow.NET libraries. The infrastructure is based on Windows Communication Foundation (WCF) for remote access, authorization, multimedia streaming and servo control. Client manual control has been implemented with the use of 3 degree-of-freedom DirectX compatible Joystick. The paper summarizes development experiences and problems concerning the use of WCF in robotics.

Solution Algorithm of Inverse Kinematics Problem for Kuka KRC3 Robots

In this paper the solution algorithm of inverse kinematics problem for KUKA KRC3 robots will be presented. Creating of this algorithm is fundamental problem of future computational intelligence for these robots. The problem of computing the joint variables corresponding a specified location of end-effector is called inverse kinematics problem. This algorithm was derived and implemented into the controller of the KUKA KRC3 robot. It allowed controlling these robots by using the smart camera NI 1742, which specifies required location of the end-effector. This required location makes it possible for the end-effector to approach a manipulation object (observed by the camera) and pick it up. The controllers of these robots have software fault, which prevents the correct cooperation with cameras. The fault was eliminated with the use of solving inverse kinematics problem, which will be presented in the this paper.

The Decreasing of 3D Position Errors in the System of Two Cameras by Means of the Interpolation Method

In this paper the original method of 3D position errors decreasing in the system of two cameras is proposed. The analysis of accuracy of determining the 3D coordinates of points on the plane template in the shape of rectangle is presented. The points lie at the corners of squares with side of 22 mm. Images of these points were obtained using Edimax IC-7100P cameras from two different points of view and analyzed. Position and orientation coordinates of cameras relatively to the reference system were calculated. Coordinates of points on the ideal image without optical distortions were determined. After reading from the image real coordinates, optical distortion model coefficients of the camera were calculated. After that, errors caused by optical distortion were determined. Coordinates read from the image were corrected and coordinates of observed points in the reference system were calculated. Next to decreasing computed 3D position errors the interpolation method was proposed. In this method the interpolation of the real coordinates of image points was used. The calculated coordinates were compared to them real values and them maximal differences were determined. Finally, the maximal position errors caused by finite dimensions of pixels were computed.

The basic component of computational intelligence for KUKA KR c3 robot

In this paper the solution algorithm of inverse kinematics problem for KUKA KR C3 robot is presented. This algorithm may be a basic component of future computational intelligence for theses robots. The problem of computing the joint variables corresponding a specified location of end-effector is called inverse kinematics problem. This algorithm was implemented into the controller of the robot. It allowed controlling these robots by using the vision system, which specifies required location of the end-effector. This required location makes it possible for the end-effector to approach a manipulation object (observed by vision system) and pick it up. These robots are equipped with several manipulator which has 6 links joined by revolute joint. First the location of end-effector in relation to the base of the manipulator were described. Next the position workspace of this robot was illustrated. The example of solutions of the inverse kinematics problem and conclusions were presented in the end of this work. In this example are the multiple solutions for singular configurations of this manipulator.

Calculation of the Location Coordinates of an Object Observed by a Camera

In this paper the algorithm Camera of the calculation of the position and orientation coordinates of the object observed by the camera is presented. This algorithm is more accurate and faster than the algorithms presented in the literature based on the minimization of the quadratic forms of errors. The camera is mounted above the technological station on which the object appears. These coordinates are calculated relative to the station frame (coordinate system associated with the technological station) or relative to the base frame (coordinate system associated with the base of robot). The orientation is described by the x-y-z fixed angles of the rotation relative to the station or the base frame. In this algorithm the perspective model of camera is used. From the image on the camera matrix sensor of three characteristic points of an object, 2D coordinates of these points are obtained. The location (position and orientation) of the object is calculated on the basis of these coordinates. The calculated location coordinates make it possible for the robot to automatically approach the object and carry out the technological operations. For example, the car’s body can constitute an object and the technological operation are to be sealing or welding.

The Algorithm Camera Computing the Object Location

In this paper the algorithm Camera of calculation of the position and orientation coordinates of the object observed by the camera is presented. The camera is mounted above the technological station on which object appears. These coordinates are calculated relative to the station frame coordinate system associated with the technological station or relative to base frame coordinate system associated with the base of robot. The orientation is described by the x-y-z fixed angles of rotation relative to station or base frame. In this algorithm the perspective model of camera is used. From the image on the camera matrix sensor of three characteristic points of the object are obtained 2D coordinates of these points. The location position and orientation of the object are calculated on the base of these coordinates. The calculated location coordinates allow the robot to automatically approach the object and carry out technological operations. For example, an object may be the car body and the technological operationi¾? sealing or welding. Creating of such algorithms is fundamental problem of computational intelligence for robots.