RUS  ENG
Full version
JOURNALS // Computer Research and Modeling // Archive

Computer Research and Modeling, 2019 Volume 11, Issue 4, Pages 631–651 (Mi crm733)

This article is cited in 2 papers

MODELS IN PHYSICS AND TECHNOLOGY

Stiffness modeling for anthropomorphic robots

D. I. Popov, A. S. Klimchik

Center for Technologies in Robotics and Mechatronics Components, Innopolis University, 1 University st., Innopolis, 420500, Russia

Abstract: In the work modeling method of anthropomorphic platforms is presented. An elastostatic stiffness model is used to determine positioning errors in the robot's lower limbs. One of the main problems in achieving a fast and stable gait are deflections caused by the flexibility in the elements of the robot. This problem was solved using virtual joint modeling to predict stiffness and deformation caused by the robot weight and external forces.
To simulate a robot in the single-support phase, the robot is represented as a serial kinematic chain with a base at the supporting leg point of contact and an end effector in the swing leg foot. In the double-support phase robot modeled as a parallel manipulator with an end effector in the pelvis. In this work, two cases of stiffness modeling are used: taking into account the compliance of the links and joints and taking into account only the compliance of joints. In the last case, joint compliances also include part of the link compliances. The joint stiffness parameters have been identified for two anthropomorphic robots: a small platform and a full-sized AR-601M.
Deflections maps were calculated using identified stiffness parameters and showing errors depending on the position of the robot end effector in the workspace. The errors in Z directions have maximum amplitude, due to the influence of the robot mass on its structure.

Keywords: stiffness modeling, elastostatic modeling, anthropomorphic platform, biped.

UDC: 004.896: 621.865

Received: 26.03.2019
Revised: 05.08.2019
Accepted: 15.08.2019

DOI: 10.20537/2076-7633-2019-11-4-631-651



© Steklov Math. Inst. of RAS, 2024