Skip to content

jhavl/swift

Repository files navigation

Swift

A Python Robotics Package QUT Centre for Robotics Open Source

PyPI version PyPI - Python Version License: MIT

Swift is a light-weight browser-based simulator built on top of the Robotics Toolbox for Python. This simulator provides robotics-specific functionality for rapid prototyping of algorithms, research, and education. Built using Python and Javascript, Swift is cross-platform (Linux, MacOS, and Windows) while also leveraging the ubiquity and support of these languages.

Through the Robotics Toolbox for Python, Swift can visualise over 30 supplied robot models: well-known contemporary robots from Franka-Emika, Kinova, Universal Robotics, Rethink as well as classical robots such as the Puma 560 and the Stanford arm. Swift is under development and will support mobile robots in the future.

Swift provides:

  • visualisation of mesh objects (Collada and STL files) and primitive shapes;
  • robot visualisation and simulation;
  • recording and saving a video of the simulation;
  • source code which can be read for learning and teaching;

Installing

Using pip

Swift is designed to be controlled through the Robotics Toolbox for Python. By installing the toolbox through PyPI, swift is installed as a dependency

pip3 install roboticstoolbox-python

Otherwise, Swift can be install by

pip3 install swift-sim

Available options are:

  • nb provides the ability for Swift to be embedded within a Jupyter Notebook
  • vision implements an RTC communication strategy allowing for visual feedback from Swift and allows Swift to be run on Google Colab

Put the options in a comma-separated list like

pip3 install swift-sim[optionlist]

From GitHub

To install the latest version from GitHub

git clone https://github.com/jhavl/swift.git
cd swift
pip3 install -e .

Code Examples

Robot Plot

We will load a model of the Franka-Emika Panda robot and plot it. We set the joint angles of the robot into the ready joint configuration qr.

import roboticstoolbox as rp

panda = rp.models.Panda()
panda.plot(q=panda.qr)

Resolved-Rate Motion Control

We will load a model of the Franka-Emika Panda robot and make it travel towards a goal pose defined by the variable Tep.

import roboticstoolbox as rtb
import spatialmath as sm
import numpy as np
from swift import Swift


# Make and instance of the Swift simulator and open it
env = Swift()
env.launch(realtime=True)

# Make a panda model and set its joint angles to the ready joint configuration
panda = rtb.models.Panda()
panda.q = panda.qr

# Set a desired and effector pose an an offset from the current end-effector pose
Tep = panda.fkine(panda.q) * sm.SE3.Tx(0.2) * sm.SE3.Ty(0.2) * sm.SE3.Tz(0.45)

# Add the robot to the simulator
env.add(panda)

# Simulate the robot while it has not arrived at the goal
arrived = False
while not arrived:

    # Work out the required end-effector velocity to go towards the goal
    v, arrived = rtb.p_servo(panda.fkine(panda.q), Tep, 1)
    
    # Set the Panda's joint velocities
    panda.qd = np.linalg.pinv(panda.jacobe(panda.q)) @ v
    
    # Step the simulator by 50 milliseconds
    env.step(0.05)

Embed within a Jupyter Notebook

To embed within a Jupyter Notebook Cell, use the browser="notebook" option when launching the simulator.

# Try this example within a Jupyter Notebook Cell!
import roboticstoolbox as rtb
import spatialmath as sm
import numpy as np
from swift import Swift

# Make and instance of the Swift simulator and open it
env = Swift()
env.launch(realtime=True, browser="notebook")

# Make a panda model and set its joint angles to the ready joint configuration
panda = rtb.models.Panda()
panda.q = panda.qr

# Set a desired and effector pose an an offset from the current end-effector pose
Tep = panda.fkine(panda.q) * sm.SE3.Tx(0.2) * sm.SE3.Ty(0.2) * sm.SE3.Tz(0.45)

# Add the robot to the simulator
env.add(panda)

# Simulate the robot while it has not arrived at the goal
arrived = False
while not arrived:

    # Work out the required end-effector velocity to go towards the goal
    v, arrived = rtb.p_servo(panda.fkine(panda.q), Tep, 1)
    
    # Set the Panda's joint velocities
    panda.qd = np.linalg.pinv(panda.jacobe(panda.q)) @ v
    
    # Step the simulator by 50 milliseconds
    env.step(0.05)