5.3: UAS Use

UAV-Based Photogrammetry on Vertical Structures (Tower)

            Photogrammetry is the science of making measurements from photographs for recovering of the exact positions of surface points or motion pathways of reference points on a moving object. It employs high-speed imaging and remote sensing to detect, measure and record complex 2D and 3D fields. According to Gruen, A. (2012), image matching is a key component of photogrammetry. Together with computer vision and image analysis, contribute to the applications of navigation, guidance, automatic surveillance, robot vision, medical image analysis and to the modelling and mapping sciences.

            In May, 2012, there was a earthquake in Emilia and Lombardy (Italy). A great number of historical buildings were seriously damaged by the shocks. In particular, most of the churches located in the southern area of Mantua’s province required restoration. The church of Santa Barbara is located in the old town center of Mantua and is one of the most important buildings in the city (Achille, C., 2015).  See highlighted building in Figure 1, 2 & 3.

                                          Figure 1. Santa Barbara Tower Bell

                                                      Figure 2. Before Quake

                                                       Figure 3. After Quake

                                                     

To restore the bell tower back to original without up-to-date drawing, photographs of 360⁰ view of the tower were necessary. If the photographs were to be taken from ground, the angles would not be accurate for imagery analysis. Some suggested to use the large crane to reach the 49 meters tower, but crane access through the small alleys is deemed a great challenge and very costly.

            The team of researchers, Achille, C. et al., (2015), opted for the choice to use a multi-copter (UAV) due to two accounts. The first consideration was the type of building: a vertical and very tall structure. UAV allows a vertical flight pattern so it permits the acquisition of vertical strips of images. Another consideration was based on the position of the building, which is in the old town centre of the city, surrounded by other buildings. For this reason it was necessary to use an easy to handle vehicle.

            The flight device had eight propellers fixed on the same number of arms, two gyroscopes for the flight control and the telemetry instruments (GPS and the barometric altimeter). The octo-copter had a flight autonomy of about five to fifteen minutes, depending on the weight loaded on board; it was equipped with LiPo batteries (16 V 4.0 Ah). The octo-copter was equipped with a reflex camera (Canon EOS 650D, APS, 18 Mega-pixel), the camera mount could tilt 90° vertically, from horizontal to zenith positions.

            The Remote Control (RC) system controls both the fly operations including camera rotation and camera trigger. The flying team included the pilot and by a photogrammetric expert able to visualize the camera view on a remotely connected monitor. This was the way to acquire images with the correct point of view and overlap.

            The most relevant step was the flight plan. It is important to define the distance from the surface, the overlap between images and, as a consequence, the trajectory. To optimize the acquisition time and reduce the number of photos, the project was optimized taking into account the camera parameters, dimension and characteristics of the building and the surroundings. The employed camera was a Canon EOS 650D with a CMOS sensor size of 5184 × 3456 pixels (22.3 × 14.9 mm) and 18 mm focal length lens. Each image was acquired with an aperture f/9 and 400 ISO. A maximum pixel size (GSD) on the object of about 3 mm was calculated, which involves an average distance of about 8 meters from the surface. An overlap of about 80% between neighbouring images was expected.

            The plan (Figure 4) was to acquire three vertical image-strips for each front, completed by two additional strips on the corners, which would permit the connection between adjacent fronts. For the acquisition of the round temple it was planned to realize three 360° flights around it, with a minimum of eight shots, completed with the same number of oblique shots from highest positions and a series of nadir photos.

            Finally, even light conditions were desired in order to have uniform colour and illumination in each image. At the same time, to avoid shadows. In this way, the photogrammetry texture and the orthophoto are uniform and similar in every part of the structure. For this reason, an overcast day was chosen to survey Santa Barbara allowing optimal light conditions.

Figure 4. Design of UAV image acquisition. In red the images for the front of the bell tower, in yellow for the round temple.

Reference:

Achille, C., Adami, A., Chiarini, S., Cremonesi, S., Fassi, F., Fregonese, L., & Taffurelli, L. (2015). UAV-based Photogrammetry and Integrated Technologies for Architectural Applications--       Methodological Strategies for the After-Quake Survey of Vertical Structures in Mantua (Italy). Sensors,  15(7), 15520-15539. doi: 10.3390/ S150715520

Gruen, A. (2012). Development and Status of Image Matching in Photogrammetry. The               Photogrammetric Record, 27(137), 36-57. doi:10.1111/j.1477-9730.2011.00671.x

2.3 - Blog: Unmanned Aerial Systems

Dear All,

            Safety is paramount in aviation. Unmanned Aerial Systems (UASs) are part of the regulatory frameworks. Pilot or controller must conform to the rules and regulations of local aviation authorities. In order to fly safely, pilot must be aware of the airspace he/ she occupies. In the U.S, there are two categories of airspace: regulatory and non-regulatory. Within these two categories there are four types: controlled, uncontrolled, special use, and other airspace (FAA, 2016). More importantly, UAS pilot must seek approval with Air Traffic Control (ATC) when flying at controlled airspace. Controlled airspace consists of:

• Class A: Airspace from 18,000 feet mean sea level (MSL) up to and including flight level (FL) 600.

• Class B: Airspace from the surface to 10,000 feet MSL surrounding the nation’s busiest airports in terms of airport operations or passenger enplanements.

• Class C: Airspace from the surface to 4,000 feet above the airport elevation (charted in MSL) surrounding those airports that have an operational control tower, are serviced by a radar approach control, and have a certain number of IFR operations or passenger enplanements.

• Class D: Class D airspace is generally airspace from the surface to 2,500 feet above the airport elevation (charted in MSL) surrounding those airports that have an operational control tower.

• Class E: If the airspace is not Class A, B, C, or D, and is controlled airspace, then it is Class E airspace.

            For general public, majority of the pilots fly theirs drones at Class G airspace, which is uncontrolled. Class G airspace extends from the surface to the base of the overlying Class E airspace. Although ATC has no authority or responsibility to control air traffic, pilots should remember there are visual flight rules (VFR) minimums which apply to Class G airspace.

            The applications of UAS increased significantly in view of the improved technologies and lower operating cost. More pilots are flying the drones within the same airspace with others UAVs or manned aircraft. Collision between two flying objects is a concern to FAA and a safety threat to the public. In order to minimize or prevent collision, the installation of sense and avoid system on UAS has become necessary.

            Angelov, P. (2012) emphasizes the sense functionalities of UAS must include detection of all hazards in the environment. It includes bad weather, wake turbulence and terrain proximity. Basic sense parameters to be considered in the design phase are:

·       The detection range of hazardous objects and avoidance maneuvering to be executed within sufficient time.

·       The field of regard, to be perceived or monitored by a sensor (camera).

·       Measurement accuracy, reliability and update rate.

·       Detection distance also depends on UAS cruise speed, turn rate, climb/ descent rates.

            After a collision threat is sensed, the UAV must perform avoidance maneuvering within structural and aerodynamic limits. These avoidance or resolution maneuvers include changes in flight trajectory, airspeed, altitude or heading. If the avoidance leads to deviation of flight path from an ATC clearance, it must be notified as soon as possible. After the conflict is solved, subsequent maneuvers must return UAV to original flight plan or to newly assigned flight path.

            Stepan. K et al. (2012) from Czech Technical University further elaborate the principle of conflict detection and resolution (CDR). The function of the collision detection and resolution system is to detect the collision and provide the resolution in the form of an evasion maneuver which is executed by UAV’s autopilot. The CDR system has five basic functions: sensing, trajectory prediction, conflict detection, conflict resolution and evasion maneuver generation.

Reference:

Angelov, P. (2012). Sense and Avoid in UAS. UK: Wiley.

FAA. (2016, April 1). Chapter 14- Airspace. US.

Stepan. K et al. (2012). Sense and Avoid Concepts: Vehicle Based SAA System. UK: Wiley.