Trajectory Optimization for Multirotors using Gurobi

Authors: Parker Lusk and Jesus Tordesillas (equally contributed)

This repo has the code implemented for the project of Underactuated Robotics (6.832) at MIT. The pdf is available here.

Flip in roll	Flip in pitch	Flip in roll with translation	Half-flip with translation

Citation

When using this code, please cite the following paper:

@techreport{luskjtorde2019,
     title = {{Trajectory Optimization for Multirotors}},
     author = {Parker Lusk and Jesus Tordesillas},
     year = {2019},
     url = {https://github.com/jtorde/uav_trajectory_optimizer_gurobi},
     institution = {Aerospace Controls Lab},
     month = {05}
}

Introduction

It uses an integrator model of any order, and solves the optimal control problem to minimize the input squared. The order of the input is specified by the user (Acceleration, Jerk, Snap, Crackle,... ). Additional position and/or attitude constraints can be specified along the trajectory.

A video of the simulation and hardware experiments is available here:

Instructions

Install Gurobi

Clone this repository:

mkdir ws && cd ws && mkdir src && cd src
git clone https://github.com/jtorde/uav_trajectory_optimizer_gurobi.git

Source the workspace:

cd ..
catkin build
source ./devel/setup.bash

Run the ROS node:

roscore
rosrun project flipper.py __ns:=SQ01s

Open Rviz and then call one of these services:

rosservice call /SQ01s/line       #Straight line
rosservice call /SQ01s/flip       #Flip in roll 
rosservice call /SQ01s/flip_pitch #Flip in pitch  
rosservice call /SQ01s/flip_trans #Flip in roll with translation
rosservice call /SQ01s/window     #Half-flip with translation

Use with an external simulator or hardware

This code can also be used with an external simulation, or run onboard a real UAV.

License

BSD 3-Clause, see License file attached