Domenico Accardo

Multi-Sensor-Based Fully Autonomous Non-Cooperative Collision Avoidance System for Unmanned Air Vehicles

This paper presents a fully autonomous multi-sensor anti-collision system for Unmanned Aerial Vehicles. This system is being developed by the Italian Aerospace Research Center in collaboration with the Department of Aerospace Engineering of the University of Naples “Federico II”. The research project is entitled TECVOL and is funded in the frame of the National Aerospace Research Program. The system prototype will be initially installed onboard a manned laboratory aircraft equipped for automatic control, therefore flight tests will verify the adequacy of attained performances for supporting fully autonomous flight. The obstacle detection and tracking function is performed by a multi-sensor configuration made up by a pulsed Ka-band radar, two visible (panchromatic and color) video cameras, two infrared video cameras, and two computers. One computer is dedicated to real time sensor fusion and communication with the radar and the flight control computer (by means of a deterministic data bus), the other is devoted to image processing. On the basis of the tracking estimates and of a Collision Avoidance Software, the flight control computer generates and follows in real-time a proper escape trajectory. In order to evaluate the performance of the collision avoidance system, numerical simulations have been performed taking into account the obstacle detection sensors’ accuracy, unmanned aircraft’s and intruder’s flight dynamics, navigation system accuracy and latencies, and collision avoidance logic. The relevant results helped to assess overall system performances and are discussed in depth.

Enhancement of the centroiding algorithm for star tracker measure refinement

Abstract The application of the hyperacuity technique to image processing of star trackers is analysed. An analytical study of the error introduced by the centroiding algorithm is presented and it is shown that a systematic contribution and a random one exist. They result from image processing assumptions and photometric measure uncertainty, respectively. Their behaviour is characterised by means of numerical simulations that are based on optics theoretical point spread functions. The latter ones take into account both defocus and diffraction effects. First, measured star position uncertainty is evaluated as a function of defocus. As a result, a criterion for optimal defocus is presented. Subsequently, an original procedure for systematic centroiding error correction by means of a backpropagation neural network is described. It is also suitable for real hardware calibration. When applied to one of the considered numerical models, the position computation accuracy is improved from 0.01 to 0.005 pixels.

Brightness-independent start-up routine for star trackers

Initial attitude acquisition by a modern star tracker is investigated here. Criteria for efficient organization of the on-board database are discussed with reference to a brightness-independent initial acquisition algorithm. Star catalog generation preprocessing is described, with emphasis on the identification of minimum star brightness for detection by a sensor based on a charge coupled device (CCD) photodetector. This is a crucial step for proper evaluation of the attainable sky coverage when selecting the stars to be included in the on-board catalog. Test results are also reported, both for reliability and accuracy, even if the former is considered to be the primary target. Probability of erroneous solution is 0.2% in the case of single runs of the procedure, while attitude determination accuracy is in the order of 0.02/spl deg/ in the average for the computation of the inertial pointing of the boresight axis.

Flight Test of a Radar-Based Tracking System for UAS Sense and Avoid

Presented here is an analysis of an extensive flight campaign aimed at characterizing peculiarities, advantages, and limitations of an obstacle detection and tracking system based on a pulse radar. The hardware and software prototypical sensing system was installed onboard an optionally piloted flying laboratory from the very light aircraft (VLA) category. Test flights with a single intruder aircraft of the same class were carried out to demonstrate autonomous noncooperative unmanned aerial system (UAS) collision avoidance capability and to evaluate the level of achievable situational awareness. First, the adopted architecture and the developed tracking algorithm are presented. Subsequently, flight data gathered in various relative flight geometries, covering chasing flights and quasi-frontal encounters, are analyzed in terms of radar performance, including detection range and range and angle measurement accuracies. The analysis describes the impact of ground echoes and navigation uncertainties, system tracking reliability, and achievable accuracy in estimation of relative position and velocity. On the basis of Global Positioning System (GPS) data gathered simultaneously with obstacle detection flight experiments, a detailed error analysis is conducted. Special emphasis is given to the validation of proposed methodology to separate between intruder and ground echoes, which is a critical aspect for light aircraft due to their limited radar cross sections (RCS) and flight altitudes. In conclusion the radar demonstrates its potential to attain adequate situational awareness, however the limits of single sensor tracking are also pointed out. Above all the negative impact of poor angular accuracy on missed detection and false alarm rates is pointed out.

An Innovative Procedure for Calibration of Strapdown Electro-Optical Sensors Onboard Unmanned Air Vehicles

This paper presents an innovative method for estimating the attitude of airborne electro-optical cameras with respect to the onboard autonomous navigation unit. The procedure is based on the use of attitude measurements under static conditions taken by an inertial unit and carrier-phase differential Global Positioning System to obtain accurate camera position estimates in the aircraft body reference frame, while image analysis allows line-of-sight unit vectors in the camera based reference frame to be computed. The method has been applied to the alignment of the visible and infrared cameras installed onboard the experimental aircraft of the Italian Aerospace Research Center and adopted for in-flight obstacle detection and collision avoidance. Results show an angular uncertainty on the order of 0.1° (rms).

Flight Performance Analysis of an Image Processing Algorithm for Integrated Sense-and-Avoid Systems

This paper is focused on the development and the flight performance analysis of an image-processing technique aimed at detecting flying obstacles in airborne panchromatic images. It was developed within the framework of a research project which aims at realizing a prototypical obstacle detection and identification System, characterized by a hierarchical multisensor configuration. This configuration comprises a radar, that is, the main sensor, and four electro-optical cameras. Cameras are used as auxiliary sensors to the radar, in order to increase intruder aircraft position measurement, in terms of accuracy and data rate. The paper thoroughly describes the selection and customization of the developed image-processing techniques in order to guarantee the best results in terms of detection range, missed detection rate, and false-alarm rate. Performance is evaluated on the basis of a large amount of images gathered during flight tests with an intruder aircraft. The improvement in terms of accuracy and data rate, compared with radar-only tracking, is quantitatively demonstrated.

Differential GNSS and Vision-Based Tracking to Improve Navigation Performance in Cooperative Multi-UAV Systems

Autonomous navigation of micro-UAVs is typically based on the integration of low cost Global Navigation Satellite System (GNSS) receivers and Micro-Electro-Mechanical Systems (MEMS)-based inertial and magnetic sensors to stabilize and control the flight. The resulting navigation performance in terms of position and attitude accuracy may not suffice for other mission needs, such as the ones relevant to fine sensor pointing. In this framework, this paper presents a cooperative UAV navigation algorithm that allows a chief vehicle, equipped with inertial and magnetic sensors, a Global Positioning System (GPS) receiver, and a vision system, to improve its navigation performance (in real time or in the post processing phase) exploiting formation flying deputy vehicles equipped with GPS receivers. The focus is set on outdoor environments and the key concept is to exploit differential GPS among vehicles and vision-based tracking (DGPS/Vision) to build a virtual additional navigation sensor whose information is then integrated in a sensor fusion algorithm based on an Extended Kalman Filter. The developed concept and processing architecture are described, with a focus on DGPS/Vision attitude determination algorithm. Performance assessment is carried out on the basis of both numerical simulations and flight tests. In the latter ones, navigation estimates derived from the DGPS/Vision approach are compared with those provided by the onboard autopilot system of a customized quadrotor. The analysis shows the potential of the developed approach, mainly deriving from the possibility to exploit magnetic- and inertial-independent accurate attitude information.

Performance Analysis and Design of an Obstacle Detection and Identification System

This paper reports the Performance Analysis and Design process for a non-collaborative anti collision sensor system to be installed onboard a High Altitude Long Endurance UAV that will be developed by the Italian Aerospace Research Center. First of all, a study on the required performances for the anti collision sensors was carried out. Requirements were critically reviewed considering flight dynamics models of the typical collision scenarios. The study included requirements on all sensor characteristics related to the anti-collision function such as resolutions coming from flying obstacle relative velocity and dimensions, field of regard, timing, and obstacle revisit rate. The requirements on sensor resolutions were discussed considering typical obstacle dimensions and flight envelopes. In addition, the dynamical correlation between sensor measurement rate and resolutions was modeled. The second step of the design process was to select sensors that fulfilled the estimated theoretical performances. The resulting sensor configuration included an active microwave anti-collision system, two visible cameras, and two infra-red sensors that allowed for increased performances in terms of angular resolution and measurement rate along with day/night and all weather capabilities. Subsequently, the system logic architecture was developed. The assigned system goal was to realize the detection, tracking, and identification of up to four obstacles that were simultaneously sensed in the Field of Regard. Standing the multiple sensor approach, sensor fusion architecture was developed for each one of the functions reported above. Multiple mode operation was determined depending on the number and on the type of sensors that could detect an obstacle at the same time. Flight tests will be carried out to demonstrate system capability with scaled performances by installing a selected set of sensors onboard a lightweight aircraft.

Satellite Angular Velocity Estimation Based on Star Images and Optical Flow Techniques

An optical flow-based technique is proposed to estimate spacecraft angular velocity based on sequences of star-field images. It does not require star identification and can be thus used to also deliver angular rate information when attitude determination is not possible, as during platform de tumbling or slewing. Region-based optical flow calculation is carried out on successive star images preprocessed to remove background. Sensor calibration parameters, Poisson equation, and a least-squares method are then used to estimate the angular velocity vector components in the sensor rotating frame. A theoretical error budget is developed to estimate the expected angular rate accuracy as a function of camera parameters and star distribution in the field of view. The effectiveness of the proposed technique is tested by using star field scenes generated by a hardware-in-the-loop testing facility and acquired by a commercial-off-the shelf camera sensor. Simulated cases comprise rotations at different rates. Experimental results are presented which are consistent with theoretical estimates. In particular, very accurate angular velocity estimates are generated at lower slew rates, while in all cases the achievable accuracy in the estimation of the angular velocity component along boresight is about one order of magnitude worse than the other two components.

Cooperative UAV navigation based on distributed multi-antenna GNSS, vision, and MEMS sensors

This paper presents an algorithm for cooperative UAV navigation that exploits vision-based sensing, standalone GPS, differential GPS among antennas embarked on different flying platforms, and measurements obtained by inertial sensors and magnetometers. Unlike other cooperative navigation approaches, the developed technique is mainly aimed at improving navigation performance in outdoor environments, either in real time or off-line. The logical architecture and the main processing steps are discussed. Then, algorithms for differential GPS/vision processing and sensor fusion for navigation state estimation are introduced. Covariance analysis is used for theoretical performance assessment. The hardware system used for concept demonstration comprises a customized quadrotor and different GPS antennas and receivers, and is briefly described. Finally, first results from experimental tests are presented. In particular, the attitude solution obtained by differential GPS and vision is compared with the estimates provided by the onboard autopilot system.