ardrone-autonomy

An autonomous flight library for the ARDrone, built on top of the node-ar-drone library. Instead of directly controlling the drone speed, you can use Autonomy to plan and execute missions by describing the path, altitude and orientation the drone must follow.

If you are a #nodecopter enthusiast, then this library will enable you to focus on higher level use cases and experiments. You focus on where you want to go, the library takes your drone there.

This work is based on the Visual Navigation for Flying Robots course.

WARNING: This is early work. Autonomous means that this library will move your drone automaticaly to reach a given target. There isn't much security in place yet, so if you do something wrong, you may have your drone fly away :-)

!! Experiment with this library in a closed/controlled environment before going in the wild !!

Features

• Extended Kalman Filter leveraging the onboard tag detection as the observation source for an Extended Kalman Filter. This provides much more stable and usable state estimate.

• Camera projection and back-projection to estimate the position of an object detected by the camera. Currently used to estimate a tag position in the drone coordinate system based on its detection by the bottom camera.

• PID Controler to autonomously control the drone position.

• Mission planner to prepare a flight/task plan and then execute it.

Planned features

• VSLAM to improve the drone localization estimates.

• Object tracking to detect and track objects in the video stream.

Mission

This module exposes a high level API to plan and execute missions, by focusing on where the drone should go instead of its low-level movements. Here is a simple example, with the drone taking off, travelling alongs a 2 x 2 meters square ane then landing.

var autonomy = require('ardrone-autonomy');
var mission  = autonomy.createMission();

mission.takeoff()
       .zero()       // Sets the current state as the reference
       .altitude(1)  // Climb to altitude = 1 meter
       .forward(2)   
       .right(2)     
       .backward(2) 
       .left(2)
       .hover(1000)  // Hover in place for 1 second
       .land();

mission.run(function (err, result)
    if (err) {
        console.trace("Oops, something bad happened: %s", err.message);
        mission.client().stop();
        mission.client().land();
    } else {
        console.log("Mission success!");
        process.exit(0);
    }
});

Mission API

mission.log(path)

Log the mission data, csv formatted, in the given file. Makes it really usefull to debug/plot the state and controller behavior.

mission.run(callback)

Execute the mission. The callback has the form function(err,result) and will be triggered in case of error or at the end of the mission.

mission.takeoff()

Add a takeoff step to the mission.

mission.forward/backward/left/right/up/down(distance)

Add a movement step to the mission. The drone will move in the given direction by the distance (in meters) before proceeding to next step. The drone will also attempt to maintain all other degrees of freedom.

mission.altitude(height)

Add a altitude step to the mission. Will climb to the given height before proceeding to next step.

mission.cw/ccw(angle)

Add a rotation step to the mission. Will turn by the given angle (in Deg) before proceeding to the next step.

mission.hover(delay)

Add a hover step to the mission. Will hover in place for the given delay (in ms) before proceeding to next step.

mission.wait(delay)

Add a wait step to the mission. Will wait for the given delay (in ms) before proceeding to next step.

mission.go(position)

Add a movement step to the mission. Will go the given position before proceeding to next step. The position is a Controller goal such as {x: 0, y: 0, z: 1, yaw: 90}.

mission.task(function)

Add a task step to the mission. Will execute the provided function before proceeding to the next step. A callback argument is passed to the function, it should be called when the task is done.

mission.zero()

Add a zeroing step to the mission. This will set the current position/orientation as the base state of the kalman filter (i.e. {x: 0, y:0, yaw:0}). If you are not using a tag as your base position, it is a good idea to zero() after takeoff.

Controller API

This module exposes a high level API to control the position. It is built using an Extended Kalman Filter to estimate the position and a PID controller to move the drone to a given target.

The easiest way to try the Controller is to play with the repl provided in the examples:

$ node examples/repl.js
// Make the drone takeoff
drone> takeoff()
// Move the drone to position (1,1)
drone> ctrl.go({x: 1, y:1});
// Climb to altitude 2 meters
drone> ctrl.altitude(2);
// Spin 90 deg to the right
drone> ctrl.cw(90);
// Go back to (0,0)
drone> ctrl.go({x:0, y:0});
// Hover in place
drone> ctrl.hover();
// Land
drone> land();

License

The MIT License

Copyright (c) 2013 by Laurent Eschenauer [email protected]

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.