Manuel Popp

Navigation Aiding by a Hybrid Laser-Camera Motion Estimator for Micro Aerial Vehicles.

Micro Air Vehicles (MAVs) equipped with various sensors are able to carry out autonomous flights. However, the self-localization of autonomous agents is mostly dependent on Global Navigation Satellite Systems (GNSS). In order to provide an accurate navigation solution in absence of GNSS signals, this article presents a hybrid sensor. The hybrid sensor is a deep integration of a monocular camera and a 2D laser rangefinder so that the motion of the MAV is estimated. This realization is expected to be more flexible in terms of environments compared to laser-scan-matching approaches. The estimated ego-motion is then integrated in the MAV’s navigation system. However, first, the knowledge about the pose between both sensors is obtained by proposing an improved calibration method. For both calibration and ego-motion estimation, 3D-to-2D correspondences are used and the Perspective-3-Point (P3P) problem is solved. Moreover, the covariance estimation of the relative motion is presented. The experiments show very accurate calibration and navigation results.

A novel guidance and navigation system for MAVs capable of autonomous collision-free entering of buildings

Micro Aerial Vehicles for autonomous explorations of hazardous areas are predestined to support emergency and rescue forces. Especially the autonomous access to buildings is highly demanding due to insufficient GNSS reception in urban terrain and narrow passageways into buildings. Thus, this paper presents a complete flight system, consisting of guidance, navigation and control subsystems. All these elements are designed to enable safe flights into buildings. The guidance subsystem is divided into two parts. The vision based guidance part is manoeuvring the MAV on an intermediate position in front of the building. The potential field based guidance part enables the MAV to fly inside the buildings without having any collisions. For that, neither any prior knowledge about the building structure, nor any maps are necessary. To provide the flight guidance with information about the actual kinematic state of the MAV an accurate and robust navigation system not depending on GNSS measurements is used. The complete system is evaluated using simulated flight data.

Model independent control of a quadrotor with tiltable rotors: IEEE/ION PLANS 2016, April 11–14, Savannah, Georgia, United States of America

In order to enable autonomous flights and to increase the field of application of Micro Aerial Vehicles (MAVs) a new quadrotor with tiltable rotors is developed. This highly non-linear system requires complex algorithms to be controlled. To gain an efficient and robust control algorithm is a key problem concerning autonomous flights. In this article a simple but robust control approach based on control allocation is presented. This approach is used in order to keep the computational costs and the level of needed system information as low as possible. By tuning the controller properly with an algorithm from the field of bionics and applying a standard downhill optimization after that the robustness and performance are increased. Simulation results show the performance of the controller in different optimization stages and display the controllers robustness in the final stage.

Automatic detection of complex shaped buildings in aerial images to support the navigation of micro aerial vehicles in urban environment

In areas with insufficient GPS reception, like in urban areas close to buildings, alternative techniques have to be used to assist the inertial navigation system. The long-term objective of this work is to use buildings, detected in camera images, as distinctive landmarks for navigating micro aerial vehicles within the aforementioned areas. This paper presents a new method to detect buildings in aerial images. To use this algorithm onboard the vehicle during the mission, it has to be fast and executed automatically without readjusting any parameters by the operator. To cover a wide range of possible application areas, no building constraints are required. Therefore a wide variation of buildings with complex shapes can be detected.