Measuring and Modelling Delays in Robot Manipulators for Temporally Precise Control Using Machine Learning

Abstract

Abstract-Latencies and delays play an important role in temporally precise robot control. During dynamic tasks in particular, a robot has to account for inherent delays to reach manipulated objects in time. The different types of occurring delays are typically convoluted and thereby hard to measure and separate. In this paper, we present a data-driven methodology for separating and modelling inherent delays during robot control. We show how both actuation and response delays can be modelled using modern machine learning methods. The resulting models can be used to predict the delays as well as the uncertainty of the prediction. Experiments on two widely used robot platforms show significant actuation and response delays in standard control loops. Predictive models can, therefore, be used to reason about expected delays and improve temporal accuracy during control. The approach can easily be used on different robot platforms.

How to achieve precise operation of a robotic manipulator on a macro to micro/nano scale

“…In reality, most control loops are in fact PI because derivative action is not used very often (Åström and Hägglund, 2001). In addition to PID control, numerous algorithms have also been developed to achieve precise operation, such as CNN (Li et al , 1998; He et al , 2016), fuzzy logic control algorithm (Yoo and Ham, 2000) and machine learning (Andersen et al , 2015) and their hybrid (Wai and Muthusamy, 2013). Because these approaches are not mature enough and their stability needs to be further verified, we just focus on the application of PID control in this section.…”

“…Although a great number of advanced control algorithms have been developed so far, such as the convolutional neural network (CNN) (Li et al , 1998; He et al , 2016), fuzzy logic control algorithm (Yoo and Ham, 2000), machine learning (Andersen et al , 2015) and their hybrid (Wai and Muthusamy, 2013), its poor adaptability and large computational load weaken their application in industry. Recently, researchers made great progress to achieve precise manipulation without sensors (Bakšys et al , 2010; Sonoda and Shimada, 2010; Kristek and Shell, 2012).…”

Real-time quintic Hermite interpolation for robot trajectory execution

“…Older versions of the Universal Robots controllers provided an unofficial C API, with very low tracking delay, as presented by Lind, Schrimpf & Ulleberg (2010). More accurate methods for measuring and classifying types of delays was presented by Andersen et al (2015).…”

Peer Review #1 of "Real-time quintic Hermite interpolation for robot trajectory execution (v0.1)"

“…Older versions of the Universal Robots controllers provided an unofficial C API, with very low tracking delay, as presented by Lind et al (2010). More accurate methods for measuring and classifying types of delays was presented by Andersen et al (2015).…”