GlobDesOpt: A Global Optimization Framework for Optimal Robot Manipulator Design

Francesco Cursi,Eric M Yeatman,Petar Kormushev,Weibang Bai

doi:10.1109/access.2022.3141660

Francesco Cursi, Eric M Yeatman + Show 2 more

Open Access

https://doi.org/10.1109/access.2022.3141660

Copy DOI

Abstract

Robot design is a major component in robotics, as it allows building robots capable of performing properly in given tasks. However, designing a robot with multiple types of parameters and constraints and defining an optimization function analytically for the robot design problem may be intractable or even impossible. Therefore black-box optimization approaches are generally preferred. In this work we propose GlobDesOpt, a simple-to-use open-source optimization framework for robot design based on global optimization methods. The framework allows selecting various design parameters and optimizing for both single and dual-arm robots. The functionalities of the framework are shown here to optimally design a dual-arm surgical robot, comparing the different two optimization strategies.

Highlights

Designing robots is a time-consuming and challenging task, which requires great expertise and knowledge on the field of application of the system
Researchers have focused on black-box optimization methods, which do not require a close form of the optimization function, nor knowledge about its derivatives
In this test GlobDesOpt was employed to optimize the design of a 4-DOF dual-arm robot, with two identical arms

Summary

Introduction

Designing robots is a time-consuming and challenging task, which requires great expertise and knowledge on the field of application of the system. Several measures can be employed to define an optimal design, such as the performance in executing a desired control task, accuracy in path tracking, and robot dexterity. Solvers like [1] using standard optimization techniques can be used for robot kinematic calibration and to improve the robot model, analytically formulating an optimization function may be challenging or even impossible, especially for highly complex robotic structures. For these reasons, researchers have focused on black-box optimization methods, which do not require a close form of the optimization function, nor knowledge about its derivatives. Hassan et al [2] employed Genetic Algorithm (GA) to optimize the design of a gripper; in [3] the optimal design of a 7DOF robot is studied using GA and considering different

Objectives

Results

Conclusion