Skip to Content

Multirotor 3D trajectory following in GPS-denied environments

 Our current multirotor controller architecture is able to perform trajectory following in indoors GPS-denied environments using a data fusion approach using measurements from the IMU, ultrasound sensors, multiple cameras and a laser range finder. The shown work is the result of multiple research stages on multirotor modeling and identification, implementation of an Extended Kalman Filter and the development of the speed planning and trajectory controller software.

Fig 1. Full controller architecture used to participate on the IMAV 2012 competition.

The videos shown at the right show different stages of the realized work to achieve the autonomous navigation capabilities of the controller architecture shown in Fig. 1. The video 1 shows the pelican autonomously following an eight-shape trajectory using only odometry measurements (no GPS or other positioning system was used). The video 2, shows the Pelican autonomously following a similar trajectory around two horizontal poles, a Particle Filter on the Hokuyo laser range finder readings is utilized. Video 3 shows part of the flight of the CVG participation on the IMAV 2012 competition (related link), the video emphasizes the role played by the particle filter of bounding the odometry cummulative estimation error. The video 4 shows work on the calibration of the Particle Filter on experimental data taken from a replica of the house of the IMAV 2012 indoors challenge. 

Fig 2. Experimental data taken from a flight performed with the Pelican quadrotor during a trajectory following task of an eight-shape trajectory. The data correspond to the Video 1 (shown at the right side of this webpage). 

This controller architecture was used by the Computer Vision Group to participate in the "Indoor Flight Dynamics – Rotory Wing MAV" challenge, obtaining two prizes: the "Best Automatic Performance - IMAV 2012" award and the second position in this challenge (related link) .

Video 1. Vertical eight-shape trajectory test

Video 2. Experimental tests for IMAV 2012 Indoors Dynamics challenge

Video 3. Particle Filter performance during IMAV 2012 Flight.

Video 4. Our Particle Filter implementation performance on experimental data.