Arun Dayal Udai

Identification of Denavit-Hartenberg Parameters of an Industrial Robot

Kinematic identification of a serial robot has been an active field of research as the need for improving the accuracy of a robot is increasing with time. Denavit-Hartenberg (DH) parameters of a serial robot, which are typically used to represent its architecture, are usually provided by its manufacturer. At times these parameters are not the same and hence they need to be identified. An analytical method proposed elsewhere was used here for identification of an industrial robot by noting the values of the point on the end-effector due to rotation of each joint, locking all other joints, were found out using singular value decomposition. The DH parameters of the robot determined using the proposed methodology, matched satisfactorily with the robot specifications. Also, the bounding volume for the joint ranges infers that a smaller measurement volume relative to the robot workspace is required thus facilitating the use of measurement devices which have smaller range of measurement.

Optimum Hip Trajectory Generation of a Biped Robot during Single Support Phase Using Genetic Algorithm

Stability of Biped walking has been an area of interest for researchers since decades. In this work an optimum hip trajectory is generated, taking its swinging foot trajectory that is a B-Spline path and physical parameters of the robot as input parameters. The objective is to minimize the deviation of zero moment point (ZMP) from the geometrical centre of supporting foot area. Genetic algorithm (GA) has emerged as a significant tool for ill faced engineering problems and can be utilized when the redundancy in solutions is exhibited. The process involves here real coded genetic algorithm (RCGA), which uses progressive search technique to find optimum hip location, for ZMP to be within certain tolerance around the most stable point. The results are simulated with a biped having 10 degrees of freedom (DOF) comprising of kinematic chain of 11 links. The variation of various joint angles can be used as reference for dynamic control of any actual robot during single support phase which is the most critical of state biped walking. The output is a more anthropomorphic and stable robot with inherent flexibility offered by a B-Spline trajectory for its modification.

Parallel active/passive force control of industrial robots with joint compliance

Active force control of an industrial robot with an end-effector force/torque sensor effectively handles compliant industrial tasks like assembly operations, surface finishing jobs, cooperative manipulation, etc. However, the robot still remains intrinsically unsafe for dynamically changing environment where the chances of the links coming in contact with the environment exists. This paper proposes a scheme for active force control at the end-effector using a six-component force/torque sensor through external force-control loop along with steady-state error compensator. In parallel, passive joint compliance was achieved by limiting the currents to joint motors based on an identified model of the robot under study. The proposed method was implemented on a KUKA KR5 ARC industrial robot and tested for passive compliance by colliding with another moving robot in its workspace. Active force control was tested to maintain a desired force on contact to demonstrate the effectiveness of the controller.

Simulation of force control algorithms for serial robots

The advantages of using force control in industrial or other type of robotic systems are well known. Study of such systems in virtual environment in the form of simulation is of great help for the design of a controller or a robot. In this paper, we show how to simulate different force control algorithms of a typical serial robot used in the industries before deciding to choose a suitable one for real implementation. This was done using an add-on to MATLAB® Simulink® known as “SimMechanics®”. Note that Simulink is an established framework for the analysis and simulation of control algorithms. Hence, our intention was to demonstrate the integration of the dynamic model of a serial robot using SimMechanics toolbox with MATLAB Simulink toolbox, e.g., Control and others.

Identifying and Updating the Kinematics of KUKA-iiwaR800 in CAD for Accurate Simulation

Kinematics and dynamics of the robot are affected by the addition of the link in the form of end-effector in the robots. An approach is presented for updating the CAD model kinematics as per customized installed robot KUKA-iiwaR800 with its gripper. KUKA-iiwaR800 is a light weight manipulator with seven degrees of freedom (DOF) which is popular in robotics research nowadays. We present a method to match the kinematic data of the installed robot with the CAD model in SimMechanics (SM) and in-house developed software RoboAnalyzer (RA). The CAD assembly of the customized robot was done in Autodesk Inventor and then imported to the Matlab environment for simulation. The validation of CAD model kinematics is done by matching the forward kinematics results from the installed robot using the log data of joint angles. After kinematic validation the model is updated in RA software with the kinematics of actual robot and the dynamic parameters as in SM. Again cycloidal trajectories were used to validate the kinematic model in RA with SM. Inverse and forward dynamics of the robot was carried out in RA using the mass and inertial properties listed in the paper.

A Quantitative Error Map Generation for Modification of Manual Socket for Below-Knee Amputee

Traditional prosthetic socket manufactured through manual plaster-casting method for lower limb prosthesis consists of geometrical errors since the shape of the cast changes due to hand impressions, pressure involved while taking out the negative cast and finally during hand sculpting of the positive cast. As the Plaster of Paris (PoP) model is further utilized for manufacturing of prosthetic limb, the error gets carried over to the prosthetic limb, which results in improper fit of the artificial limb and ultimately wounding the amputees residual limb, particularly at the stress-prone knee. In the present work, an error map generation using reverse engineering (RE) technique has been proposed which assists in rectifying the manually prepared PoP model. The correction technique is quantitative and requires less hand-sculpting skills of the prosthetist. With this process, the accuracy increases to the level of a product manufactured by CNC/RP method and at a comparatively much cheaper rate.

Force/Position Control of 3 DOF Delta Manipulator with Voice Coil Actuator

Parallel manipulators are widely used in the industries for several applications. Due to its precision in motion as well as its robustness, parallel manipulators have proved its advantage over serial manipulators. In this paper, a 3DOF parallel manipulator is presented and force control of the manipulator is demonstrated. The proposed manipulator uses a direct drive voice coil arc actuators to achieve compliance required for human-robot interaction or soft mechanical manipulations. Its implementation in the proposed delta manipulator is discussed in the paper. The paper has discussed a unique method of controlling position as well as the force at the end-effector of the delta manipulator. The method used in making the manipulator compliant does not need an explicit force sensor and is convenient to implement. The method is inexpensive and works satisfactorily in a human interactive environment which is demonstrated through experiments discussed in the paper. The proposed design finds its application in robot-assisted assembly, surface finishing, cooperative manipulation, haptics etc.

Identification and control of NAO humanoid robot to grasp an object using monocular vision

NAO is a humanoid robot which is developed by Aldebaran robotics, a French robotic company having its headquarters in Paris. Various works related to monocular vision for hand eye coordination in NAO robot has been done which are explained in this paper. Camera calibration is done to determine parameters defining relationship between reference 3D coordinate system and camera coordinate frame, which are combined with transformation and parameters related to camera. Control of robotic arm is a bit complex, it requires complete dynamics study and its implementation. Though it is a wide area, some of the works are done on its arm control using different methods. Accurate control requires identification of D-H parameters and forward and inverse kinematics of NAO arm which needs to have knowledge about degrees of freedom (DOF). Adaptive Neuro Fuzzy Inference System(ANFIS) is used for inverse kinematics calculation. Future works in this direction could be implementing controllers which would help to reduce the stability problem.

A framework for CAD-based offline depth-map preparation for automated assembly tasks

This paper presents an intuitive offline depth-map generation tool that can be applied for any generic assembly task using industrial robots. Such depth-maps are commonly required to be generated for any new robotic assembly in hand. Once the assembly of the peg and hole is designed in the Computer Aided Design (CAD) platform, the software framework proposed in this paper is used to generate the depth-map automatically.