Biomimetic Energy-Based Humanoid Gait Design (original) (raw)

An energy efficient gait for a Nao robot

Benelux Artificial Intelligence Conference (BNAIC), 2013

The gait of humans is often assumed to be the most energy efficient way of walking. Srinivasan and Ruina [18] confirm this hypothesis using a simple model in which the human is a point mass with straight legs that can change in length during a step. Their results show that the inverted pendulum walk is the most energy efficient gait. The question is whether this result also holds for humanoid robots. This paper investigate what is the most energy efficient gait for a humanoid robot such as the Nao, and what the corresponding control policy is that needs to be implemented. To answer these questions, first, the model of Srinivasan and Ruina is adapted for humanoid robots, and is used to study the energy consumption of different gaits. The model assumes a gait with dynamic stability and assumes that the torque on the knee joint provides the main contribution to the energy consumption of a gait. The former assumption implies that no energy is needed to remain stable. The latter assumption is confirmed by an experiment with a humanoid robot, namely Nao. Based on experiments with this idealize model, a gait that minimizes the energy consumption is identified. A controller for the new gait is implemented and is evaluated on a Nao robot. In the future, this controller will be the basis of an intelligent controller that can adapt to varying circumstances.

An Energy Efficient Dynamic Gait for a Nao Robot

IEEE International Conference on Autonomous Robot Systems and Competitions (ICARSC), 2014

—This paper presents a framework to generate energy efficient dynamic human-like walk for a Nao humanoid robot. We first extend the inverted pendulum model with the goal of finding an energy efficient and stable walking gait. In this model, we propose a leg control policy which utilizes joint stiffness control. We use policy gradient reinforcement learning to identify the optimal parameters of the new gait for a Nao humanoid robot. We successfully test the control policy in a simulator and on a real Nao robot. The test results show that the new control policy realizes a dynamic walk that is more energy efficient than the standard walk of Nao robot.

Using the Generalized Inverted Pendulum to Generate Less Energy-Consuming Trajectories for Humanoid Walking

Archive of Mechanical Engineering, 2016

This paper proposes an analysis of the effect of vertical position of the pivot point of the inverted pendulum during humanoid walking. We introduce a new feature of the inverted pendulum by taking a pivot point under the ground level allowing a natural trajectory for the center of pressure (CoP), like in human walking. The influence of the vertical position of the pivot point on energy consumption is analyzed here. The evaluation of a 3D Walking gait is based on the energy consumption. A sthenic criterion is used to depict this evaluation. A consequent reduction of joint torques is shown with a pivot point under the ground.

Dynamic Lateral Stability for an Energy Efficient Gait

Benelux Artificial Intelligence Conference (BNAIC), 2014

This paper presents an energy efficient dynamically stable gait for a Nao humanoid robot. In previous work we identified a dynamically stable and energy efficient gait in the sagittal or walking direction of a Nao robot. This gait proved to be more energy efficient than the standard gait, provided by the manufacturer. Dynamic stability in the lateral direction was not addressed. Lateral stability was handled by full stiffness of the joint in lateral direction. In this paper we report on adding dynamic lateral stability. We do not yet incorporate feedback of sensors. This implies that the gait is only suited for flat horizontal surfaces that some lateral joint stiffness is needed in the implementation on the Nao.

A Bipedal Walking Robot with Efficient and Human-Like Gait

Proceedings of the 2005 IEEE International Conference on Robotics and Automation, 2005

Here we present the design of a passivedynamics based, fully autonomous, 3-D, bipedal walking robot that uses simple control, consumes little energy, and has human-like morphology and gait. Design aspects covered here include the freely rotating hip joint with angle bisecting mechanism; freely rotating knee joints with latches; direct actuation of the ankles with a spring, release mechanism, and reset motor; wide feet that are shaped to aid lateral stability; and the simple control algorithm. The biomechanics context of this robot is discussed in more detail in [1], and movies of the robot walking are available at Science Online and http://www.tam.cornell.edu/∼ruina/powerwalk.html. This robot adds evidence to the idea that passive-dynamic approaches might help design walking robots that are simpler, more efficient and easier to control.

Teo: Full-Size Humanoid Robot Design Powered by a Fuel Cell System

Cybernetics and Systems, 2012

This article deals with the design of the full-size humanoid robot TEO, an improved version of its predecessor Rh-1. The whole platform is conceived under the premise of high efficiency in terms of energy consumption and optimization. We will focus mainly on the electromechanical structure of the lower part of the prototype, which is the main component demanding energy during

A Bio-Inspired Approach to the Realization of Sustained Humanoid Motion

2012

Abstract This paper overviews some author's biomechanical inspiration for the development of an approach which enables the realization of bipedal artificial motion. First, we introduce the notion of dynamic balance, which is a basic prerequisite for the realization of any task by humanoids. Then, as ground reference points, important indicators of a humanoid's state were introduced and discussed. Particular attention was paid to ZMP, which is the most important indicator of robot dynamic balance.

Can Walking Be Modeled in a Pure Mechanical Fashion

Intelligent Autonomous Systems 15, 2018

The aim of this paper is to investigate the role of some mechanical quantities in the challenging task to make a robot walking or running. Because the upright posture of an humanoid is the main source of instability, the maintenance of the equilibrium during locomotion requires the gait-controller to deal with a number of constraints, such as ZMP, whose dynamical satisfactions prevent the humanoid from an harmful fall. Walking humanoids are open systems heavily interacting with a perturbing environment and the rapid loss of mechanical energy could be an hallmark of instability. In this paper we shall show how certain dimensionless parameters could be useful to design the walking gait of a bipedal robot.

New joint design to create a more natural and efficient biped

Applied Bionics and Biomechanics, 2009

This paper presents a human-oriented approach to design the mechanical architecture and the joint controller for a biped robot. Starting from the analysis of the human lower limbs, we figured out which features of the human legs are fundamental for a correct walking motion, and can be adopted in the mechanical design of a humanoid robot. We focus here on the knee, designed as a compliant human-like knee instead of a classical pin-joint, and on the foot, characterised by the mobility and lightness of the human foot. We implemented an elastic actuator, with a simple position control paradigm that sets the joint stiffness in real time, and developed the basic controller. Results in simulation are discussed. In our approach the robot gains in adaptability and energetic efficiency, which are the most challenging issues for a biped robot.