Skip to content

Control and motion planning for an aerial robot with visual odometry

Notifications You must be signed in to change notification settings

harshg99/control_sensing_estimation_quadrotor

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

control_sensing_estimation_quadrotor

Introduction

The goal of the project was to develop a complete planning and control system with state estimation using visual inertial odometry (VIO) for a quadrotor, enabling it to fly from an initial point to a goal point efficiently and without collision with obstacles or crashing. With perfect information of the state of the system aggressive trajectories and controllers were feasible. With state estimation, however, aggressive trajecto- ries designed in prior projects are not feasible due to tracking errors in position and attitude of the quadrotor introduced by the Visual Inertial Odometry(VIO) that is sensitive to themotion of the quad-rotor. Thus, this project aims to outline the changes made in the trajectory generation and controls of the system, in order for the quadrotor to perform well under the constraints introduced by state estimation.

Controller performance

image

Path Planning

image

Visual Inertial Odometry Estimation Performance

image

Trajectory optimisation vs Path taken by quadrotor

image

About

Control and motion planning for an aerial robot with visual odometry

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages