Marcin Odelga

Obstacle detection, tracking and avoidance for a teleoperated UAV

In this paper, we present a collision-free indoor navigation algorithm for teleoperated multirotor Unmanned Aerial Vehicles (UAVs). Assuming an obstacle rich environment, the algorithm keeps track of detected obstacles in the local surroundings of the robot. The detection part of the algorithm is based on measurements from an RGB-D camera and a Bin-Occupancy filter capable of tracking an unspecified number of targets. We use the estimate of the robots velocity to update the obstacles state when they leave the direct field of view of the sensor. The avoidance part of the algorithm is based on the Model Predictive Control approach. By predicting the possible future obstacles states, it filters the operator commands to prevent collisions. The method is validated on a platform equipped with its own computational unit, which makes it self-sufficient in terms of external CPUs.

Control of 3 DOF helicopter: A novel autopilot scheme based on adaptive sliding mode control

This paper presents a novel autopilot for a 3D helicopter. From desired trajectories defined by the user for elevation and travel angles, the autopilot is computing the desired trajectory of the pitch angle. Furthermore, the autopilot allows to decouple the system and to define “virtual” inputs in order to separately design controllers for each attitude angle. Travel and elevation controllers are based on adaptive version of sliding mode control: this class of controllers keeps the robustness feature of sliding mode while reducing the well-known drawback of such control approach, the chattering, thanks to the online adaptation of the controller gain. The proposed autopilot is evaluated on an experimental set-up.

A Setup for multi-UAV hardware-in-the-loop simulations

In this paper, we present a hardware-in-the-loop simulation setup for multi-UAV systems. With our setup we are able to command the robots simulated in Gazebo, a popular open source ROS-enabled physical simulator, using computational units that are embedded on our quadrotor UAVs. Hence, we can test in simulation not only the correct execution of algorithms, but also the computational feasibility directly on the robots hardware. In addition, since our setup is inherently multi-robot, we can also test the communication flow among the robots. We provide two use cases to show the characteristics of our setup.

A fully actuated quadrotor UAV with a propeller tilting mechanism: Modeling and control

Equipped with four actuators, quadrotor Unmanned Aerial Vehicles belong to the family of underactuated systems. The lateral motion of such platforms is strongly coupled with their orientation and consequently it is not possible to track an arbitrary 6D trajectory in space. In this paper, we propose a novel quadrotor design in which the tilt angles of the propellers with respect to the quadrotor body are being simultaneously controlled with two additional actuators by employing the parallelogram principle. Since the velocity of the controlled tilt angles of the propellers does not appear directly in the derived dynamic model, the system cannot be static feedback linearized. Nevertheless, the system is linearizable at a higher differential order, leading to a dynamic feedback linearization controller. Simulations confirm the theoretical findings, highlighting the improved motion capabilities with respect to standard quadrotors.

Efficient real-time video stabilization for UAVs using only IMU data

While some unmanned aerial vehicles (UAVs) have the capacity to carry mechanically stabilized camera equipment, weight limits or other problems may make mechanical stabilization impractical. As a result many UAVs rely on fixed cameras to provide a video stream to an operator or observer. With a fixed camera, the video stream is often unsteady due to the multirotors movement from wind and acceleration. These video streams are often analyzed by both humans and machines, and the unwanted camera movement can cause problems for both. For a human observer, unwanted movement may simply make it harder to follow the video, while for computer algorithms, it may severely impair the algorithms intended function. There has been significant research on how to stabilize videos using feature tracking to determine camera movement, which in turn is used to manipulate frames and stabilize the camera stream. We believe, however, that this process could be greatly simplified by using data from a UAVs on-board inertial measurement unit (IMU) to stabilize the camera feed. In this paper we present an algorithm for video stabilization based only on IMU data from a UAV platform. Our results show that our algorithm successfully stabilizes the camera stream with the added benefit of requiring less computational power.

Nonlinear Modeling and Control of a 3 DOF Helicopter

This paper presents a novel autopilot for a 3D helicopter. From desired trajectories defined by the user for elevation and travel angles, the autopilot is computing the desired trajectory of the pitch angle. Furthermore, the autopilot allows to decouple the system and to define “virtual” inputs in order to separately design controllers for each attitude angle. Travel and elevation controllers are based on adaptive version of super-twisting algorithm: this class of controllers keeps the robustness feature of sliding mode while reducing the well-known drawback of such control approach, the chattering, thanks to the online adaptation of the controller gain.© 2012 ASME

A control architecture for physical human-UAV interaction with a fully actuated hexarotor

Physical human-UAV interaction (PHUI) with Unmanned Aerial Vehicles (UAV) has many possible applications, and a few recent works [1], [2] have shown that it is possible to perform PHUI with a standard quadrotor. However, the intrinsic underactuation of quadrotors may hinder the interaction task and also cause safety issues. In this paper, we present an admittance control-based scheme to perform PHUI with a fully actuated UAV. The system also benefits from the robustness provided by our fully actuated UAV implementation of the adaptive super twisting controller (ASTC). While validating our system in simulation, we also show the superior performance of the fully actuated UAV with respect to a standard quadcopter.