Enrico Villagrossi

Optimal robot dynamics local identification using genetic-based path planning in workspace subregions

Methods for dynamic calibrations of Industrial Robots (IR) are increasing their importance in many applications because of high performances attained by model-based control strategies. Most of known state-of-the-art methods aim at modeling robots along the complete workspace, often affecting the identified parameters with loss of physical meaning (e.g. negative inertia values) and requiring a wide exploration of the workspace both in term of joint positions and velocities (accelerations). Actually, many IR tasks require dynamic accuracy in limited portion of the workspace and commonly display mild dynamics. Local identification of dynamics parameters in task conditions could therefore increase the predictive capability of the model for that operation. This work proposes the use of a parametric-description of trajectories in Cartesian space, corresponding to the standard industrial path-description as a series of via-points in most of programming languages. The identification of the optimal exciting Cartesian trajectory in a local sub-region of the workspace is made by a genetic algorithm over the template trajectory description. The use of an IR real interpolator allows to match computational and task execution conditions.

A general analytical procedure for robot dynamic model reduction

The identification of the dynamic model of a robotic manipulator represents a fundamental step for designing high performance model-based controllers. Despite the huge number of works presented on this topic, the symbolic dynamic model reduction (i.e., the identification of the set of parameters observable through the measure of joint torques and positions) still remain a challenging task, characterized from tailored solutions, adapted from time to time to specific families of mechanisms. The work here presented, introduces an automatic and analytical reduction of the dynamic model, based on a multi-dimensional Fourier series decomposition of the dynamic equations. The procedure enables to obtain symbolically the base dynamic parameters (BP) starting from a given kinematic structure. The Fourier based model reduction can be applied indifferently both to open- and closed-chain kinematics. A simulated example shows the effectiveness of the proposed algorithm.

Robot dynamic model identification through excitation trajectories minimizing the correlation influence among essential parameters

Robot dynamics is commonly modeled as a linear function of the robot kinematic state from a set of dynamic parameters into motor torques. Base parameters (i.e. the set of theoretically demonstrated linearly-independent parameters) can be reduced to a subset of “essential” parameters by eliminating those that are negligible with respect to their contribution in motor torques. However, generic trajectories, if not properly defined, couple the contribution of such essential parameters into the motor torques, actually reducing the estimation accuracy of the dynamics parameters. The work presented here introduces an index for evaluating correlation influence among essential parameters along an executed trajectory. Such index is then exploited for an optimal search of excitatory patterns consistent with the kinematical coupling constraints. The method is experimentally compared with the results achievable by one of the most popular IRs dynamic calibration method.

On robot dynamic model identification through sub-workspace evolved trajectories for optimal torque estimation

Model-based control are affected by the accuracy of dynamic calibration. For industrial robots, identification techniques predominantly involve rigid body models linearized on a set of minimal lumped parameters that are estimated along excitatory trajectories made by suitable/optimal path. Although the physical meaning of the estimated lumped models is often lost (e.g. negative inertia values), these methodologies get remarkably results when well-conditioned trajectories are applied. Nonetheless, such trajectories have usually to span the workspace at large, resulting in an averagely fitting model. In many technological tasks, instead, the region of dynamics applications is limited, and generation of trajectories in such workspace sub-region results in different specialized models that should increase the predictability of local behavior. Besides this consideration, the paper presents a genetic-based selection of trajectories in constrained sub-region. The methodology places under optimization paths generated by a commercial industrial robot interpolator, and the genes (i.e. the degrees-of-freedom) of the evolutionary algorithms corresponds to a finite set of few via-points and velocities, just like standard motion programming of industrial robots. Remarkably, experiments demonstrate that this algorithm design feature allows a good matching of foreseen current and the actual measured in different task conditions.

A human mimicking control strategy for robotic deburring of hard materials

ABSTRACT This paper deals with the use of an industrial robot (IR) for the deburring of hard material items (i.e. cast iron items). The control strategies introduced in this paper aim to mimic the human behaviour during the manual deburring. On the basis of force feedback, provided from a 1-axis load cell, the nominal deburring trajectory is optimised and deformed making multiple repetitions. The deburring trajectory is repeated until completing the nominal deburring path. The removal of thin layers of materials allows the robot to operate at high feed rates avoiding spindle stall and without exciting elastics effects on the mechanical structure of the system. Furthermore, a method to automatically detect the force changepoints, related to the presence of a burr, without tuning force thresholds, is discussed. The human mimicking control strategy is compared with a standard industrial approach demonstrating a reduction of the task cycle time and an improvement of the finishing quality.