0
Research Papers

Rapid Manufacture of Novel Variable Impedance Robots

[+] Author and Article Information
Alexander Enoch

Institute of Perception,
Action and Behaviour,
School of Informatics,
University of Edinburgh,
Edinburgh EH8 9YL, UK
e-mail: a.m.enoch@sms.ed.ac.uk

Sethu Vijayakumar

Institute of Perception,
Action and Behaviour,
School of Informatics,
University of Edinburgh,
Edinburgh EH8 9YL, UK
e-mail: sethu.vijayakumar@ed.ac.uk

Manuscript received August 1, 2014; final manuscript received April 8, 2015; published online August 18, 2015. Assoc. Editor: Aaron M. Dollar.

J. Mechanisms Robotics 8(1), 011003 (Aug 18, 2015) (11 pages) Paper No: JMR-14-1190; doi: 10.1115/1.4030388 History: Received August 01, 2014

Variable stiffness and variable damping can play an important role in robot movement, particularly for legged robots such as bipedal walkers. Variable impedance also introduces new control problems, since there are more degrees of freedom to control, and the resulting robot has more complex dynamics. In this paper, we introduce novel design and fabrication methodologies that are capable of producing cost effective hardware prototypes suitable for investigating the efficacy of impedance modulation. We present two variable impedance bipedal platforms produced using a combination of waterjet cutting and 3D printing, and a novel fused deposition modeling (FDM) 3D printing based method for producing hybrid plastic/metal parts. We evaluate walking trajectories at different speeds and stiffness levels.

FIGURES IN THIS ARTICLE
<>
Copyright © 2016 by ASME
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Fig. 1

BLUE and miniBLUE, robots capable of mechanically varying the dynamics of their joints, and 3D printed part with waterjet cut aluminum embedded inside during printing

Grahic Jump Location
Fig. 2

Schematic of the variable stiffness transmission. The motor attached to the input link drives the intermediate arm which, through a pair of compression springs, delivers torque to the output link. A second motor adjusts the position of the compression springs, and hence the distance r in order to change the stiffness.

Grahic Jump Location
Fig. 3

Torque and stiffness curves for the variable stiffness mechanism in the knees and ankles of BLUE

Grahic Jump Location
Fig. 4

BLUE: a planar biped

Grahic Jump Location
Fig. 5

(a) miniBLUE: a biped with 2DOF hips and torso; (b) a variable stiffness pod which attaches to the side of miniBLUE, forming the series elastic element in the drive. A motor on the pod drives the central leadscrew, adjusting the position of the load bearing springs and thereby changing the joint stiffness; and (c) miniBLUE joint architecture: damping motor connectivity in parallel with variable stiffness pod.

Grahic Jump Location
Fig. 6

Design and evaluation of the compliant feet of miniBLUE: (a) miniBLUE's foot, (b) FEA of miniBLUE foot arch, and (c) simulated and measured deflection of the arch

Grahic Jump Location
Fig. 7

Design and evaluation of the 3D printed compliant arch in the foot of BLUE: (a) BLUE's foot with 3D printed arch, (b) FEA of BLUE foot arch, and (c) simulated and measured deflection of the arch

Grahic Jump Location
Fig. 8

FDM printing of hybrid part, encasing waterjet cut aluminum: (a) before first metal piece, (b) after first metal piece, (c) channel for sensor and recess for crossbar, and (d) all metal pieces in place

Grahic Jump Location
Fig. 9

The electronics layout of BLUE. Modular control boards linked via Ethernet network.

Grahic Jump Location
Fig. 10

Peak ground reaction forces (GFR) on the phalanges of the foot versus longitudinal arch stiffness, for three different walking gaits. Generally, a softer arch reduces peak GRF on the toes.

Grahic Jump Location
Fig. 11

Walking trajectory playback on the robot, both in simulation and on the actual hardware. (a) Simulation and (b) BLUE.

Grahic Jump Location
Fig. 12

Joint angles, equilibrium angles, stiffness setting, and deflection from equilibrium in the knee joint of BLUE during squatting: (a) simulation of squatting while changing stiffness and (b) hardware squatting while changing stiffness

Grahic Jump Location
Fig. 13

Tuning the behavior of the robot to conform more closely to a desired output trajectory. Over three iterations, the output of the robot is pushed toward the desired result with a proportional gain.

Grahic Jump Location
Fig. 14

Walking at a different stiffness levels, for two different speeds. (a) At very slow speeds, the task effectively becomes point-to-point motion, and we observe a higher energy usage in the robot at lower stiffness levels. (b) At faster speeds, decreasing the stiffness level can decrease energy usage.

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In