Bayesian Optimization for Whole-Body Control of High-Degree-of-Freedom Robots Through Reduction of Dimensionality

Yuan, Kai; Chatzinikolaidis, Iordanis; Li, Zhibin

doi:10.1109/lra.2019.2901308

Cited by 39 publications

(31 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For the sake of completeness, it is worth mentioning other works applying Bayesian optimization in robotics applications. In particular, (Drieß, Englert, & Toussaint, 2017) employs constrained BO to select three force-controller parameters for combined position/interaction tasks; (Cully, Clune, Tarapore, & Mouret, 2015) describes a trial-and-error algorithm that allows robots to adapt their behavior in presence of damage; (Yuan, Chatzinikolaidis, & Li, 2019) proposes a methodology to achieve automatic tuning of optimal parameters for whole-body control algorithms, iteratively learning the parameters of sub-spaces from the whole high-dimensional parametric space through interactive trials. Our contribution differs from the works mentioned above since it is tailored to industrial manipulators with unknown dynamics and adopting a widely-used control architecture, which relies on a simple feedback linerizator for dynamics compensation, a feedforward action, and PID controllers at the joint level.…”

Section: Paper Contributionmentioning

confidence: 99%

Robot control parameters auto-tuning in trajectory tracking applications

Roveda

Forgione

Piga

2020

Control Engineering Practice

View full text Add to dashboard Cite

Autonomy is increasingly demanded to industrial manipulators. Robots have to be capable to regulate their behavior to different operational conditions, adapting to the specific task to be executed without requiring high time/resource-consuming human intervention. Achieving an automated tuning of the control parameters of a manipulator is still a challenging task, which involves modeling/identification of the robot dynamics. This usually results in an onerous procedure, both in terms of experimental and data-processing time. This paper addresses the problem of automated tuning of the manipulator controller for trajectory tracking, optimizing control parameters based on the specific trajectory to be executed. A Bayesian optimization algorithm is proposed to tune both the low-level controller parameters (i.e., the equivalent link-masses of the feedback linearizator and the feedforward controller) and the high-level controller parameters (i.e., the joint PID gains). The algorithm adapts the control parameters through a data-driven procedure, optimizing a user-defined trajectory-tracking cost. Safety constraints ensuring, e.g., closed-loop stability and bounds on the maximum joint position error are also included. The performance of proposed approach is demonstrated on a torquecontrolled 7-degree-of-freedom FRANKA Emika robot manipulator. The 25 robot control parameters (i.e., 4 link-mass parameters and 21 PID gains) are tuned in less than 130 iterations, and comparable results with respect to the FRANKA Emika embedded position controller are achieved. In addition, the generalization capabilities of the proposed approach are shown exploiting the proper reference trajectory for the tuning of the control parameters.

show abstract

Section: Paper Contributionmentioning

confidence: 99%

Robot control parameters auto-tuning in trajectory tracking applications

Roveda

Forgione

Piga

2020

Control Engineering Practice

View full text Add to dashboard Cite

show abstract

“…A final weight of approximately 6 kg and the inertias for each part are calculated and added to the urdf file. The robot has a total of 32 DoF (degrees of freedom); Two of these 32 joints are purely passive [8], as shown in table 1.…”

Section: Foot Height (Fh)mentioning

confidence: 99%

Adam - A Virtual Humanoid Robot Model Based on Open Source Approach

França¹,

Oliveira²,

Souza³

et al. 2020

Blucher Engineering Proceedings

View full text Add to dashboard Cite

The Robotics has been driven by the possibilities of using robots to assist human beings. The design of a humanoid robotic system aims to obtain a machine adapted to the human environment and that can perform tasks to assist people with greater adaptability and ease of execution. This article presents the humanoid robot Adam, who will have an open architecture platform for simulations. The prototype will be used as a testing ground and thus promote the advances in the most diverse techniques in robotics and artificial intelligence. Adam is a virtual humanoid robot, and this article is about it and its virtual environment.

show abstract

“…In order to execute robot movements optimally, some studies set or define multiple reference trajectories for the robot, and optimize controller parameters. In [17], [18], [19] for instance, the authors parameterize hard or soft priorities for multiple trajectories in a QP controller and posteriorly optimize them w.r.t. to different cost functions related to the humanoid kinematics or dynamics.…”

Section: Related Workmentioning

confidence: 99%

Humanoid Whole-Body Movement Optimization from Retargeted Human Motions

Gomes

Radhakrishnan

Penco

et al. 2019

2019 IEEE-RAS 19th International Conference on Humanoid Robots (Humanoids)

View full text Add to dashboard Cite

Motion retargeting and teleoperation are powerful tools to demonstrate complex whole-body movements to humanoid robots: in a sense, they are the equivalent of kinesthetic teaching for manipulators. However, retargeted motions may not be optimal for the robot: because of different kinematics and dynamics, there could be other robot trajectories that perform the same task more efficiently, for example with less power consumption. We propose to use the retargeted trajectories to bootstrap a learning process aimed at optimizing the whole-body trajectories w.r.t. a specified cost function. To ensure that the optimized motions are safe, i.e., they do not violate system constraints, we use constrained optimization algorithms. We compare both global and local optimization approaches, since the optimized robot solution may not be close to the demonstrated one. We evaluate our framework with the humanoid robot iCub on an object lifting scenario, initially demonstrated by a human operator wearing a motion-tracking suit. By optimizing the initial retargeted movements, we can improve robot performance by over 40%.

show abstract

Bayesian Optimization for Whole-Body Control of High-Degree-of-Freedom Robots Through Reduction of Dimensionality

Cited by 39 publications

References 25 publications

Robot control parameters auto-tuning in trajectory tracking applications

Robot control parameters auto-tuning in trajectory tracking applications

Adam - A Virtual Humanoid Robot Model Based on Open Source Approach

Humanoid Whole-Body Movement Optimization from Retargeted Human Motions

Contact Info

Product

Resources

About