Quick Start

In this page, we first introduce the architecture, then we show how to run the basic applications developed with ManiWare.

Prerequisites

We assume a working installation of ManiWare and Pybullet is available (see the installation page). Moreover, we assume the reader to be familiar with the basic concepts of ManiWare, Pybullet and Python.

ManiWare introduction

ManiWare is a middleware developed by Zhiqin Cheng and Jinlin Chen from Department of Computing, The Hong Kong Polytechnic University.

The aim of ManiWare is to provide an easy-to-use middleware that provides a team-level programming abstraction and the manipulator-level plugin mechanism for programming and building manipulator applications.

The package is already equipped with basic team-level cooperation modules, and manipulator-level functional components, so that developers can start to build the prototype on basic robot models such as mobile manipulator and fixed-base manipulator for different applications.

We decompose the software stack into two layers to reduce the design complexity, i.e., the team-level cooperation layer and the manipulator-level motion control layer. The architecture of ManiWare is shown as below:

Please refer to API Documentation for a detailed description of the package functionalities.!

How to launch an application

We provide three case studies to show the basic features of ManiWare. Users can launch the applications directly.

Case study 1: Cooperative pick-and-place

cd example
python3 exp_pick_place.py

Case study 2: Cooperative object collection

cd example
python3 exp_collection.py

Case study 3: Dynamic Reconfiguration

cd example
python3 exp_dyn_reconf.py

Developers can take the above case studies as references to build different applications.