Quadrotor simulation

This project contains:

• Informed RRT* to plan a collision free path from start to goal position in an N-dimensional Map.
• A minimum snap trajectory generator. The segment times between consecutive waypoints are also optimised.
• Nonlinear geometric controller for aggressive trajectory tracking and hover controller
• Quadrotor Simulator

Dependencies

• scipy, numpy, matplotlib, rtree-linux

Usage

Trajectory generation and control example

from TrajGen import trajGenerator, Helix_waypoints
from Quadrotor import QuadSim
import controller

waypoints = Helix_waypoints(5)

#Generate trajectory through waypoints
traj = trajGenerator(waypoints,max_vel = 10,gamma = 1e6)

#initialise simulation with given controller and trajectory
Tmax = traj.TS[-1]
des_state = traj.get_des_state
sim = QuadSim(controller,des_state,Tmax)

#run simulation
sim.run()

With Path Planning

python3 runsim.py

find out more in PathPlanning/README.md

Future Work

• Adding disturbance to quadrotor model
• Collision checking after trajectory generation

References

• Polynomial Trajectory Planning for Aggressive Quadrotor Flight in Dense Indoor Environments - Charles Richter, Adam Bry, and Nicholas Roy
• Trajectory Generation and Control for Quadrotors - Daniel Mellinger and Vijay Kumar
• Motion Planning as Search - Russ Tedrake