Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Joint-Space Control of a Structurally Elastic Humanoid Robot

Connor W. Herron, Christian Runyon, Isaac Pressgrove, Benjamin C. Beiter, Bhaben Kalita, Alexander Leonessa

Abstract

In this work, the joint-control strategy is presented for the humanoid robot, PANDORA, whose structural components are designed to be compliant. As opposed to contemporary approaches which design the elasticity internal to the actuator housing, PANDORA's structural components are designed to be compliant under load or, in other words, structurally elastic. To maintain the rapid design benefit of additive manufacturing, this joint control strategy employs a disturbance observer (DOB) modeled from an ideal elastic actuator. This robust controller treats the model variation from the structurally elastic components as a disturbance and eliminates the need for system identification of the 3D printed parts. This enables mechanical design engineers to iterate on the 3D printed linkages without requiring consistent tuning from the joint controller. Two sets of hardware results are presented for validating the controller. The first set of results are conducted on an ideal elastic actuator testbed that drives an unmodeled, 1 DoF weighted pendulum with a 10 kg mass. The results support the claim that the DOB can handle significant model variation. The second set of results is from a robust balancing experiment conducted on the 12 DoF lower body of PANDORA. The robot maintains balance while an operator applies 50 N pushes to the pelvis, where the actuator tracking results are presented for the left leg.

Joint-Space Control of a Structurally Elastic Humanoid Robot

Abstract

In this work, the joint-control strategy is presented for the humanoid robot, PANDORA, whose structural components are designed to be compliant. As opposed to contemporary approaches which design the elasticity internal to the actuator housing, PANDORA's structural components are designed to be compliant under load or, in other words, structurally elastic. To maintain the rapid design benefit of additive manufacturing, this joint control strategy employs a disturbance observer (DOB) modeled from an ideal elastic actuator. This robust controller treats the model variation from the structurally elastic components as a disturbance and eliminates the need for system identification of the 3D printed parts. This enables mechanical design engineers to iterate on the 3D printed linkages without requiring consistent tuning from the joint controller. Two sets of hardware results are presented for validating the controller. The first set of results are conducted on an ideal elastic actuator testbed that drives an unmodeled, 1 DoF weighted pendulum with a 10 kg mass. The results support the claim that the DOB can handle significant model variation. The second set of results is from a robust balancing experiment conducted on the 12 DoF lower body of PANDORA. The robot maintains balance while an operator applies 50 N pushes to the pelvis, where the actuator tracking results are presented for the left leg.

Paper Structure

This paper contains 13 sections, 18 equations, 16 figures, 1 table.

Table of Contents

  1. Introduction
  2. Structurally Elastic Humanoid
  3. Humanoid Control Overview
  4. Joint-Space Control
  5. Structural Elasticity Modeling
  6. Actuator Force Controller
  7. Virtual Impedance
  8. Leaky Integration and Kinematics
  9. Efficient Continuous to Digital Transformation
  10. Results and Discussion
  11. Pendulum Control on Elastic Actuator Testbed
  12. Robust Balancing on PANDORA
  13. Conclusion

Figures (16)

  • Figure 1: Full rendering of PANDORA humanoid robot (left) with the lower-body fully built and operational (right)
  • Figure 2: Comparison between structural elasticity from PANDORA (left) and conventional elasticity approaches (right) within the actuators.
  • Figure 3: PANDORA's control hierarchy is broken down into high and low-level controllers utilizing motor current, quadrature and absolute encoder, and force sensor feedback from an actuator/joint pair.
  • Figure 4: Overview of the Joint-Space Controller that uses leaky integration and kinematics to identify actuator objectives and impedance and force Control to achieve those trajectories.
  • Figure 5: Overview of lower body measurement axes. These measurements relate to the kinematic transform from the joint and actuator spaces.
  • ...and 11 more figures