Differences

This shows you the differences between two versions of the page.

--- en:localisation_et_commande_embarquee_d_un_drone_en_utilisant_la_vision_stereoscopique [2016/04/07 17:22] – castillo
+++ en:localisation_et_commande_embarquee_d_un_drone_en_utilisant_la_vision_stereoscopique [2016/04/07 17:34] (current) – castillo
@@ Line 15: / Line 15: @@
 Grant from __Mexicain Government__ \\
 Location: Heudiasyc \\
-Date PhD finished: November, 2011 \\
+Date PhD finished: **November 28th, 2011** \\
 \\
-__Current position__: **MdC at GIPSA Lab**\\
+__Current position__: **MdC at GIPSA Lab since 2013**\\
 \\
@@ Line 25: / Line 25: @@
 == Abstract ==
-The present document addresses, theoretically and experimentally, the most relevant topics
+Visual servoing is a control approach based on visual information. In this thesis,
-for Unmanned Aerial Vehicles (UAVs) in autonomous and semi-autonomous navigation.
+visual servoing schemes are proposed to control a quadrotor and an octarotor applied
-According with the multidisciplinary nature of the studied problems, a wide range of
+to positioning and navigation task. Concerning the quadrotor we use a hierarchical
-techniques and theories are covered in the fields of robotics, automatic control, computer
+control scheme whose inner-loop (fast dynamic) focuses on attitude dynamics, while
-science, computer vision and embedded systems, among others.\\
+outer-loop (slow dynamics) deals with translational dynamics.\\
-As part of this thesis, two different experimental platforms were developed in order to
-explore and evaluate various theories and techniques of interest for autonomous navigation.
-The first prototype is a quadrotor specially designed for outdoor applications and was fully
-developed in our lab. The second testbed is composed by a non expensive commercial
-quadrotor kind AR. Drone, wireless connected to a ground station equipped with the Robot
-Operating System (ROS), and specially intended to test computer vision algorithms and
-automatic control strategies in an easy, fast and safe way.\\
-In addition, this work provides an study of data fusion techniques looking to enhance
+Also, a nonlinear controller based on separated saturations for a quadrotor is
-the UAVs pose estimation provided by commonly used sensors. Two strategies are evaluated
+proposed to stabilize it attitude. The linear position and velocity of the rotorcraft are
-in particular, an Extended Kalman Filter (EKF) and a Particle Filter (PF). Both estimators
+obtained by using a vision-based algorithm via a monocular caméra. The dynamic
-are adapted for the system under consideration, taking into account noisy measurements
+model of the quadrotor is presented using the Newton-Euler formalism.
-of the UAV position, velocity and orientation. Simulations show the performance of
+In other vision system, two cameras are used to estimate the translational position
-the developed algorithms while adding noise from real GPS (Global Positioning System)
+and velocity of the vehicle. Position was obtained using a frontal camera looking
-measurements.\\
+at a target placed on a wall. Quadrotor velocity was estimated using a camera pointing
+vertically downwards running an optical flow algorithm. Experimental tests
+showed that the quadrotor performed well at hover flight using the proposed vision
+based control system.\\
-Safe and accurate navigation for either autonomous trajectory tracking or haptic teleoperation
+– **Quadrotor vision-based**\\
-of quadrotors is presented as well. A second order Sliding Mode (2-SM) control
-algorithm is used to track trajectories while avoiding frontal collisions in autonomous flight.
-The time-scale separation of the translational and rotational dynamics allows to design position
-controllers by giving desired references in the roll and pitch angles, which is suitable
-for quadrotors equipped with an internal attitude controller. The 2-SM control allows to
-add robustness to the closed-loop system and attenuates the chattering effect. A Lyapunov
-based analysis probes the system stability. Vision algorithms are employed to estimate the
-pose of the vehicle using only a monocular SLAM (Simultaneous Localization and Mapping)
-fused with inertial measurements. Distance to potential obstacles is detected and
-computed using the sparse depth map from the vision algorithm. For teleoperation tests,
-a haptic device is employed to feedback information to the pilot about possible collisions,
-by exerting opposite forces. The proposed strategies are successfully tested in real-time
-experiments, using a low-cost commercial quadrotor. \\
-Also, conception and development of a Micro Aerial Vehicle (MAV) able to safely
+The same system was used to estimate the 3D position of the quadrotor over
-interact with human users by following them autonomously, is achieved in the present
+a trajectory using vanishing points. The performance of the vision and control
-work. Once a face is detected by means of a Haar cascade classifier, it is tracked applying
+algorithms has been tested in a real application by a quadrotor tracking a line
-a Kalman Filter (KF), and an estimation of the relative position with respect to the face
+painted in a wall. Similarly the velocity estimation is obtained using an optic
-is obtained at a high rate. A linear Proportional Derivative (PD) controller regulates the
+flow algorithm. The estimated position and velocity information obtained
-UAV’s position in order to keep a constant distance to the face, employing as well the extra
+from the vision system is combined with the angular rates and displacements
-available information from the embedded UAV’s sensors. Several experiments were carried
+of the inertial measurement unit to compute the control inputs. It has been
-out through different conditions, showing good performance even under disadvantageous
+shown that the proposed control scheme achieves the tracking objective of
-scenarios like outdoor flight, being robust against illumination changes, wind perturbations,
+the visual reference.\\
-image noise and the presence of several faces on the same image.
-\\
+– **Octarotor vision-based** \\
+In this thesis, it is presented a visual feedback a control of an octarotor
+using image-based visual servoing (IBVS) with stereo vision. Autonomous
+control of an UAV requires a precise measurements and/or estimation of the
+vehicle’s pose and also the knowledge of its surrounding environment. In
+order to control the orientation and the position of flying robot with respect
+to a target, we propose to use a navigation system based on binocular vision
+system combined with inertial sensors. This combination of sensors, allows
+us to get a complete characterization of the state of aerial vehicle. In other
+words, using the stereo vision system we are able to estimate the UAV’s
+D position, while from the inertial sensors we obtain the orientation of
+rotorcraft. A semi-embedded navigation system combining stereo vision with
+inertial information is proposed.\\
+The hierarchical control approach is appropriate to stabilize the 6DOF dynamics
+of the quadrotor, it takes advantage of the time scale separation between rotational
+(fast) and translational (slow) dynamics. For this reason, despite the lower frequency
+rate of vision-based measurements is able to stabilize in real-time the quadrotor
+translational dynamics. This combination of measurement strategies has many advantages
+because one works very well at low speeds (vision system) and the other at
+high speeds (inertial sensors). Both work at different sample rate. Taking advantage
+of this property we have obtained a simplified dynamical model of the rotorcraft.
+This model is given by six independent double integrators which have been stabilized
+using proportional-derivative (PD) control. The real-time experiments have
+shown an acceptable performance of the flying machine applying the control law
+and sensing system proposed.\\
-Finally, this thesis deals with the problem of implementing a safe and fast transportation
+An embedded control system for the mini rotorcraft is implemented. The control
-system using an UAV kind quadrotor with a cable suspended load. The objective consists
+is validated by experimental tests. Experimental results show that the implementation
-in transporting the load from one place to another, in a fast way and with minimum
+of the control law on an embedded control system is satisfactory for autonomous
-swing in the cable. An Interconnection and Damping Assignment-Passivity Based Control
+hovering in indoors and outdoors with light or no wind. Real time experiences are
-(IDA-PBC) for the system of interest is successfully implemented and tested in real time
+developed to validate the performance of navigation systems proposed. This work
-experiments. The use of this control technique avoids the use of an extra sensor for the
+highlights the potential of the computer vision based position control strategies for
-swing angle measurement. An experimental platform composed by a quadrotor equipped
+UAV.
-with a rigid rod with two degrees of freedom in the joint, is developed. UAV navigation is
-achieved via monocular vision.
 \\
 \\
 \\

UMR CNRS 7253

Site Tools

Differences

Page Tools

User Tools