This work presents a probabilistic model for learning robot tasks from human demonstrations using kinesthetic teaching. The difference with respect to previous works is that a complete state of the robot is used to obtain a consistent representation of the dynamics of the task. The learning framework is based on hidden Markov models and Gaussian mixture regression, used for coding and reproduci...

In this paper, two adaptive iterative learning control schemes, proposed by A. Tayebi [2004, Automatica, 40(7), 1195–1203], are tested experimentally on a five-degrees-of-freedom (5-DOF) robot manipulator CATALYST5. The control strategy consists of using a classical PD feedback structure plus an additional iteratively updated term designed to cope with the unknown parameters and disturbances. T...

In this paper we introduce two new posture-dependent performance indices for 6dof UPS and 3-dof RPR parallel manipulators. One is based on an object-oriented metric in the workspace (end-effector dependent) and the other one takes the angular velocities of the passive joints into consideration (end-effector independent). Both newly defined indices are invariant under rigid-body motions and simi...

A novel concept of neural force position control in cartesian space NFC was developed and applied The NFC concept for a DOF industrial robot with DOF sensor forces torques is based on a cycle time of only msec NFC features include Sensor and trajectory data processing in cartesian space Learned mapping operations for force kinematics and dynamics with neural networks Singularity robust ness in ...

A new robot toolbox for dynamics simulation based on MATLAB Graphical User Interface (GUI) is developed for educational purposes. It is built on the previous version named as ROBOLAB performing only the kinematics analysis of robot manipulators. The toolbox presented in this paper provides interactive real-time simulation and visualization of the industrial robot manipulator’s dynamics based on...

trajectory using genetic algorithms (GA) is presented. The efficiency of the proposed approach, without loss of generality, is demonstrated through a simulation carried out on a planar 6-DOF robot manipulator. The approach benefits from two key features. First, the method of data representation which guarantees the satisfaction of joints angle limits, and second the conversion of considered mod...