Gen Endo

Learning CPG-based Biped Locomotion with a Policy Gradient Method: Application to a Humanoid Robot

In this paper we describe a learning framework for a central pattern generator (CPG)-based biped locomotion controller using a policy gradient method. Our goals in this study are to achieve CPG-based biped walking with a 3D hardware humanoid and to develop an efficient learning algorithm with CPG by reducing the dimensionality of the state space used for learning. We demonstrate that an appropriate feedback controller can be acquired within a few thousand trials by numerical simulations and the controller obtained in numerical simulation achieves stable walking with a physical robot in the real world. Numerical simulations and hardware experiments evaluate the walking velocity and stability. The results suggest that the learning algorithm is capable of adapting to environmental changes. Furthermore, we present an online learning scheme with an initial policy for a hardware robot to improve the controller within 200 iterations.

Experimental Studies of a Neural Oscillator for Biped Locomotion with QRIO

Recently, there has been a growing interest in biologically inspired biped locomotion control with Central Pattern Generator (CPG). However, few experimental attempts on real hardware 3D humanoid robots have yet been made. Our goal in this paper is to present our achievement of 3D biped locomotion using a neural oscillator applied to a humanoid robot, QRIO. We employ reduced number of neural oscillators as the CPG model, along with a task space Cartesian coordinate system and utilizing entrainment property to establish stable walking gait. We verify robustness against lateral perturbation, through numerical simulation of stepping motion in place along the lateral plane. We then implemented it on the QRIO. It could successfully cope with unknown 3mm bump by autonomously adjusting its stepping period. Sagittal motion produced by a neural oscillator is introduced, and then overlapped with the lateral motion generator in realizing 3D biped locomotion on a QRIO humanoid robot.

A Biologically Inspired Biped Locomotion Strategy for Humanoid Robots: Modulation of Sinusoidal Patterns by a Coupled Oscillator Model

Biological systems seem to have a simpler but more robust locomotion strategy than that of the existing biped walking controllers for humanoid robots. We show that a humanoid robot can step and walk using simple sinusoidal desired joint trajectories with their phase adjusted by a coupled oscillator model. We use the center-of-pressure location and velocity to detect the phase of the lateral robot dynamics. This phase information is used to modulate the desired joint trajectories. We do not explicitly use dynamical parameters of the humanoid robot. We hypothesize that a similar mechanism may exist in biological systems. We applied the proposed biologically inspired control strategy to our newly developed human-sized humanoid robot computational brain (CB) and a small size humanoid robot, enabling them to generate successful stepping and walking patterns.

Modulation of simple sinusoidal patterns by a coupled oscillator model for biped walking

We show that a humanoid robot can step and walk using simple sinusoidal desired joint trajectories with their phase adjusted by a coupled oscillator model. We use the center of pressure location and velocity to detect the phase of the lateral robot dynamics. This phase information is used to modulate the desired joint trajectories. We applied the proposed control approach to our newly developed human sized humanoid robot and a small size humanoid robot developed by Sony, enabling them to generate successful stepping and walking patterns

Poincaré-Map-Based Reinforcement Learning For Biped Walking

We propose a model-based reinforcement learning algorithm for biped walking in which the robot learns to appropriately modulate an observed walking pattern. Via-points are detected from the observed walking trajectories using the minimum jerk criterion. The learning algorithm modulates the via-points as control actions to improve walking trajectories. This decision is based on a learned model of the Poincaré map of the periodic walking pattern. The model maps from a state in the single support phase and the control actions to a state in the next single support phase. We applied this approach to both a simulated robot model and an actual biped robot. We show that successful walking policies are acquired.

A passive weight compensation mechanism with a non-circular pulley and a spring

We propose a new weight compensation mechanism with a non-circular pulley and a spring. We show the basic principle and numerical design method to derive the shape of the non-circular pulley. After demonstration of the weight compensation for an inverted/ordinary pendulum system, we extend the same mechanism to a parallel five-bar linkage system, analyzing the required torques using transposed Jacobian matrices. Finally, we develop a three degree of freedom manipulator with relatively small output actuators and verified that the weight compensation mechanism significantly contributes to decrease static torques to keep the same posture within manipulators work space.

Quadruped walking robots at Tokyo Institute of Technology

In this article, the design principle of the leg driving mechanism to minimize energy loss and maximize output power is discussed. We will also introduce the gait control methods implemented in our previous quadruped walking robots. Finally, we will survey most of the prototype models of our quadruped walking robots.

Study on Roller-Walker - Adaptation of characteristics of the propulsion by a leg trajectory -

Roller-walker is a leg-wheel hybrid mobile robot using a passive wheel equipped on the tip of each leg. The passive wheel can be transformed into sole mode by rotating ankle roll joints when roller-walker walks on rough terrain. This paper describes adaptation of characteristics of the propulsion by a leg trajectory in the case of wheeled locomotion. Firstly, the authors demonstrate that Roller-Walker could achieve high-speed propulsion and slope climbing propulsion by simply changing parameters of the leg trajectory on the hardware experiments. Secondly, an asymptotic parameter tuning method is introduced to perform specified velocity on the different surfaces with different friction. The method is evaluated in numerical simulations. The results suggest that the method allows the Roller-Walker to have a function similar to an automatic transmission of a usual car.

A framework for learning biped locomotion with dynamical movement primitives

This article summarizes our framework for learning biped locomotion using dynamical movement primitives based on nonlinear oscillators. Our ultimate goal is to establish a design principle of a controller in order to achieve natural humanlike locomotion. We suggest dynamical movement primitives as a central pattern generator (CPG) of a biped robot, an approach we have previously proposed for learning and encoding complex human movements. Demonstrated trajectories are learned through movement primitives by locally weighted regression, and the frequency of the learned trajectories is adjusted automatically by a frequency adaptation algorithm based on phase resetting and entrainment of coupled oscillators. Numerical simulations and experimental implementation on a physical robot demonstrate the effectiveness of the proposed locomotion controller. Furthermore, we demonstrate that phase resetting contributes to robustness against external perturbations and environmental changes by numerical simulations and experiments.

Study on Roller-Walker - Energy efficiency of Roller-Walk -

Roller-Walker is a leg-wheel hybrid mobile robot using a passive wheel equipped on the tip of each leg. The passive wheel can be transformed into sole mode by rotating ankle roll joint when Roller-Walker walks on rough terrain. This paper discusses energy efficiency of locomotion in wheeled mode. We define a leg trajectory to produce forward straight propulsion and discuss the relationships between the parameters of the leg trajectory and energy efficiency of the propulsion using numerical simulator. We find optimum parameter sets where optimization criteria is specific resistance. Then we carried out hardware experiments and empirically derived experimental specific resistance. We show that wheeled locomotion has 8 times higher energy efficiency than ordinary crawl gait. Finally we compare the specific resistance of Roller-Walker with other walking robots described in the literatures.