Omnidirectional static walking of a quadruped robot

Ma, Shugen; Tomiyama, T.; Wada, Hideo

doi:10.1109/tro.2004.835448

Cited by 91 publications

(59 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There is direct correlation between the two phases: the displacement of the main-body along the motion direction in the stance phase has to match the leg advancement along the very same direction during the swing phase. According to the actuators, only Static Gaits and Marginally Static Gaits [10,12] have been developed for Warugadar, in particular a variation the crawl gait, illustrated in Fig. 3.…”

Section: Hardware: Mechanical Structure Actuators and Sensorsmentioning

confidence: 99%

A New Distributed Software Architecture Within the Pyro Environment for a Quadruped Robot

2009

View full text Add to dashboard Cite

The robot platforms are usually composed with a bunch of different elements coming from different technologies, therefore merging and handling information, from sensors to actuators, is a complex problem. Our aim is obtain a controller of a legged robot walking on unstructured environments and to develop robust algorithms to coordinate the movement according to a variety of requirements. We built a small 4-legged autonomous robot, named Warugadar, endowed with monocular camera, piezoelectric contact sensors and ZigBee transceivers, thought to be modular and scalable. In fact it is possible to change hardware components without deeply modifying its software architecture. Each component of the robot (data acquisition board for sensors, camera, motor board) is stand-alone and shares information and receives commands from a brain unit on a remote computer powered by Pyro (Python Robotics). Pyro is an open-source robotics toolkit written in Python for exploring topics in AI and Robotics, which introduces generic robot abstractions that are uniform across a number of robot platform regardless of their size and morphology. In this way it is possible to control the behaviors of Warugadar.

show abstract

Section: Hardware: Mechanical Structure Actuators and Sensorsmentioning

confidence: 99%

A New Distributed Software Architecture Within the Pyro Environment for a Quadruped Robot

2009

View full text Add to dashboard Cite

show abstract

“…Much of the research has focused on static walking (moving one leg at a time, thereby keeping the robot statically stable) including statically stable gaits capable of omnidirectional path following [19]. In addition, there has been much work on designing static gaits that can navigate irregular terrain [6,5,11,18].…”

Section: A Quadruped Locomotionmentioning

confidence: 99%

Learning omnidirectional path following using dimensionality reduction

Kolter¹,

Ng²

2007

Robotics: Science and Systems III

View full text Add to dashboard Cite

Abstract-We consider the task of omnidirectional path following for a quadruped robot: moving a four-legged robot along any arbitrary path while turning in any arbitrary manner. Learning a controller capable of such motion requires learning the parameters of a very high-dimensional policy, a difficult task on a real robot. Although learning such a policy can be much easier in a model (or "simulator") of the system, it can be extremely difficult to build a sufficiently accurate simulator. In this paper we propose a method that uses a (possibly inaccurate) simulator to identify a low-dimensional subspace of policies that spans the variations in model dynamics. This subspace will be robust to variations in the model, and can be learned on the real system using much less data than would be required to learn a policy in the original class. In our approach, we sample several models from a distribution over the kinematic and dynamics parameters of the simulator, then formulate an optimization problem that can be solved using the Reduced Rank Regression (RRR) algorithm to construct a low-dimensional class of policies that spans the major axes of variation in the space of controllers. We present a successful application of this technique to the task of omnidirectional path following, and demonstrate improvement over a number of alternative methods, including a hand-tuned controller. We present, to the best of our knowledge, the first controller capable of omnidirectional path following with parameters optimized simultaneously for all directions of motion and turning rates.

show abstract

“…In contrast to the CPG-based method, the WT-based method has great advantages in the foot WT arbitrarily designing and planning [4,[20][21][22], such as excellent capability of over-obstacle, smooth motion, smaller mechanical shock, lower energy consumption, etc. [4,6,23,24].…”

Section: Introductionmentioning

confidence: 99%

A Bio-Inspired Control Strategy for Locomotion of a Quadruped Robot

Zeng

Yang

et al. 2018

Applied Sciences

View full text Add to dashboard Cite

Abstract:In order to effectively plan the robot gaits and foot workspace trajectory (WT) synchronously, a novel biologically inspired control strategy for the locomotion of a quadruped robot based on central pattern generator-neural network-workspace trajectory (CPG-NN-WT) is presented in this paper. Firstly, a foot WT is planned via the Denavit-Hartenberg (D-H) notation and the inverse kinematics, which has the advantages of low mechanical shock, smooth movement, and sleek trajectory. Then, an improved central pattern generator (CPG) based on Hopf oscillators is proposed in this paper for smooth gait planning. Finally, a neural network is designed and trained to convert the CPG output to the preplanned WT, which can make full use of the advantages of the CPG-based method in gait planning and the WT-based method in foot trajectory planning simultaneously. Furthermore, virtual prototype simulations and experiments with a real quadruped robot are presented to validate the effectiveness of the proposed control strategy. The results show that the gait of the quadruped robot can be controlled easily and effectively by the CPG with its internal parameters; meanwhile, the foot trajectory meets the preplanned WT well.

show abstract

Omnidirectional static walking of a quadruped robot

Cited by 91 publications

References 15 publications

A New Distributed Software Architecture Within the Pyro Environment for a Quadruped Robot

A New Distributed Software Architecture Within the Pyro Environment for a Quadruped Robot

Learning omnidirectional path following using dimensionality reduction

A Bio-Inspired Control Strategy for Locomotion of a Quadruped Robot

Contact Info

Product

Resources

About