Motion Signals With Velocity Jumps: Velocity Estimation Employing Only Quantized Position Data

Abstract

DOI to the publisher's website. • The final author version and the galley proof are versions of the publication after peer review. • The final published version features the final layout of the paper including the volume, issue and page numbers. Link to publication General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal. If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the "Taverne" license above, please follow below link for the End User Agreement:

Robot control for simultaneous impact tasks via QP based reference spreading

“…with threshold ǫ. In real-life robotic applications, switching will occur based on impact detection, for example through jump-aware filtering [29], as is used in the experimental validation of reference spreading in [19].…”

Predicting the Post-Impact Velocity of a Robotic Arm via Rigid Multibody Models: an Experimental Study

“…Proposed quantitative comparison procedure. Summarizing, we propose the following method to quantitatively compare experimental post-impact data with post-impact velocity predictions derived from a rigid multi-body impact map: 1) Identify the impact time t i in the experimental data by looking at, e.g., sharp variations of joint encoder data (either visually as we did in this work or by the use of automatic methods such as those described in [33] and references therein); 2) Extract the impact robot configuration q(t i ) and corresponding pre-impact joint velocity q− (t i ) and compute the rigid-robot post-impact joint velocity estimate ∆ q+ (t i ) employing the impact map (5); 3) Use (nonlinear) least squares fitting or frequencydomain-based procedures on the signal q(t) to separate the affine (=constant plus linear) response from the oscillatory damped response ("sum of eigenmodes") over the interval [t i , t i + T s ] where vibrations are observed (with T s the settling time). Employ the affine part to construct the virtual rigid-robot post-impact velocity q+ (t i ); 4) Evaluate the (relative and absolute) error between ∆ q+ (t i ) and q+ (t i ) for impacts occurring at different postures and velocities, to quantify the general accuracy of the post-impact velocity estimation (considering the virtual rigid-robot post-impact velocity q+ (t i ) the ground truth obtained from experiments).…”

Reference Spreading: Tracking Performance for Impact Trajectories of a 1DoF Setup

“…For performing the trajectory tracking experiments in Section IV-C, we use the JA filtering method introduced in [26] to estimate position q and velocity v from the encoder data. The method is employed because it not only accurately estimates state in the presence of velocity jumps but also includes an impact detection feature.…”

“…When a second-order low-pass filter is used for estimating velocity, as opposed to a method tailored to discontinuous velocity signals [26], the inevitable lag in the response to an impact causes large (velocity) errors right after impact. These fictitious errors are fed back to the system via feedback which, in turn, results in larger tracking errors.…”

“…This choice is made to demonstrate that RS can successfully be implemented in situations with limited sensory information. The technique known as jump aware (JA) filtering [26] is employed for estimating position and velocity as well as for detecting impacts. On the other hand, the goal of [25] was to provide an experimental proof of principle, the objective of this brief is to analyze the control strategy from a performance and robustness perspective.…”