Unmanned Aircraft Systems (english version), Appunti di Ingegneria Aerospaziale

Scarica Unmanned Aircraft Systems (english version) e più Appunti in PDF di Ingegneria Aerospaziale solo su Docsity! UNMANNED AIRCRAFT SYSTEM LECTURE 1 What is an Unmanned Aircraft System? Which are its main components? The field of unmanned aircraft systems (UAS) is very broad and it covers myriad missions and systems types. The definition of unmanned vehicle given by the United States department of Defense (DoD) is the following: “A powered vehicle that does not carry a human operator, can be operated autonomously or remotely, can be expandable or recoverable, and can carry a lethal or non-lethal payload. Unmanned vehicles are the primary component of unmanned systems. Ballistic or semi-ballistic vehicle, cruise missiles, artillery projectiles, torpedoes, mines, satellites, and unattended sensors (with no form of propulsion) are not considered unmanned vehicles.” The unmanned system must include a flying vehicle, there has to be a form of propulsion (except for gliders) and the unmanned aircraft must be capable of controlled and sustainable flight. The system comprises a number of subsystems which include the aircraft, payloads, control stations, communication subsystems, aircraft launch and recovery subsystems. Whatever they are called, unmanned aircraft systems have an airborne component that performs at least one mission role without a pilot onboard. Some of the common terms used to refer to unmanned air vehicles are drone, remotely piloted vehicle (RPV), unmanned aerial vehicle (UAV), remotely piloted aircraft (RPA), remotely piloted helicopter (RPH). ICAO has defined some of these names: -Unmanned Aircraft System (UAS): an aircraft and its associated elements which are operated with no pilot onboard; -Remotely Piloted Aircraft System (RPAS): a set of configurable elements consisting of a remotely- piloted aircraft, its associated remote pilot stations, the required command and control links and any other system elements required at any point during flight operation; -Remotely Piloted Aircraft (RPA): an aircraft where the pilot is not onboard. Which are the basic implications of the absence of the onboard pilot? (slide 17) The absence of the onboard pilot (and therefore a fly by wireless conduction of the vehicle) results in a situation awareness reduction because no feedback comes from the pilot; as a consequence, another effect is the vulnerability of command and control structure and the system is automatic to autonomous operations. Automation in the cockpit is and advantage because it requires no personnel, but when the automatics don’t do what they’re supposed to do and no pilot is present to intervene, it becomes a big problem. Moreover, “unmanned” is a misnomer, because in reality the system is remotely controlled by a human, and so a human is actually always present, even if not physically onboard, and this presence is also needed when something happens and a responsible must be found. Describe the basic UAS motivations, and some typical dull, dirty or dangerous missions. Unmanned aircraft have some advantages compared to manned aircraft. UAS offer smaller potential size, because they are smaller and lighter and with greater stealth capabilities (and so they are less detectable by other systems). They are versatile, have the ability of high or low inspections and come in various sizes, providing a wide selection to suit a variety of inspection needs. UAS might provide greater performance than manned aircraft and they are more flexible: they can take photos, capture videos, take thermal images, transmit data and other functions. Another significant reason for UAS introduction in the military field is that they eliminate the need for workers to physically access hostile environments; drones can easily access difficult-to-reach areas. In addition, operating costs of UAS are reduced: maintenance costs, fuel costs, hangarage costs are lower. Insurance requirements for drones are minimal and mission based, hence with lower costs (though this depends on individual circumstances). Finally, UAS allow to save time and money: drones can minimize the cost and time requires to erect ladders, access towers, aerial lifts and other heavy equipment. UAS are used for dull, dirty and dangerous missions (3D), mainly in the military but also in the civilian field. With dull missions we refer to military (and civilian) applications such as extended surveillance, communication relay, air sampling, which can be a dull experience for aircrew, with many hours spent on watching without relief, leading to a loss of concentration and therefore a loss of mission effectiveness. For example, in the long duration surveillance case, the UAS with high resolution colour video, low light level TV, thermal imaging cameras or radar scanning can be more effective and even cheaper than manned aircrew to operate in such case. The ground-based operators can be readily relieved in a shift work pattern. Dirty missions involve flight through contaminated air that would be harmful to humans; contaminants can include radiation, biological or chemical agents. The UAS might be required to sample or measure the contents of the air or perform other missions in this environment. Detoxification of the aircraft is also easier in the case of UAV. For military dangerous applications, such as suppression of enemy air defense or reconnaissance of heavily defended areas, the detection rate o a manned aircraft is likely to exceed that of a UAV, which is smaller and provided with greater stealth techniques. The UAV operators are under no personal threat and can concentrate specifically and more effectively on the task; as a result, the UAV offers a greater probability of mission success without the risk of losing aircrew. Some dangerous applications in the civilian field are forest fire control, operations in extreme weather conditions, flight inside volcano craters, etc. actuators commanded by the autopilot can directly drive control surfaces and engines. In some cases, OPA suffer flight performance penalties relative to dedicated unmanned aircraft. Examples of recent OPA are Boeing-Gulfstream G-550 (fixed wing) and SW4 “SOLO” RUAS (rotary wing). LECTURE 4 Which are the basic requirements for UAS navigation systems (see slide 5)? Small systems require integrated flight control systems comprising navigation sensors. General basic requirements for UAS navigation systems derive from flight control, attitude and pointing accuracy and needs related to navigation error. Flight control is related to both ground devices (Ground Control Station-GCS) and onboard devices. UAVs require hardware and software elements allowing the aircraft to be controlled remotely either directly by a pilot or autonomously by onboard computers. UAV flight dynamics are highly variable and non-linear, so maintaining attitude and stability may require continuous computations and readjustments of the aircraft system. Describe the different types of UAS landing aids. Whether an unmanned aircraft lands on a traditional runway, captures in a net or other recovery system is employed, landing aids are often necessary. The landing point of the unmanned aircraft must be intercepted within an acceptable tolerance and this accuracy is often greater than what could be achieved by GPS alone. Landing aids are also requested to enable autonomous landing on fixed or moving platforms. The main adopted landing aids are RF (radio frequency) based, ground- laser based, relative GPS positioning, AGL (above ground level) sensors, vision systems. RF based methods generally use a narrow beam to guide the unmanned aircraft to the touchdown point. The signal may be encoded with ground-based correction information. Some examples are Sierra Nevada Tactical Automatic Landing System (TALS) and UAV Common Automatic Recovery System V2-35 GHz. Ground laser systems are based on the same working principle of RF based methods and they adopt a laser beam and a retroreflection system onboard the UAV. Relative GPS positioning methods determine the relative position between two GPS receivers. Communications must occur between the two receivers by means of a data link. Code phase differential GPS and real-time kinematic (carrier phase differential GPS) methods are commonly used. AGL sensors provide estimate of the height of the unmanned aircraft above the local surface: radar, laser and acoustic altimeters are used. The uses for AGL sensors can be ground collision avoidance, navigation by feature mapping and support of the final stages of landing. The most significant distinction between laser altimeters and radar altimeters relies upon order of magnitude smaller footprint that results from the narrow beam of optical radiation. The radar has a pulse-limited footprint resulting from a wavefront curvature and pulse length; on the other hand, in laser altimeters pulse-limited and beam-limited footprints are identical. The result is a very high quality measurement for individual laser pulses and that’s why laser altimeters are more accurate than radar in good visibility situations. Moreover, laser altimeter signals don’t introduce electromagnetic compatibility problems. Machine vision systems can interpret the scene ahead of the unmanned aircraft to assist landing; vision-based systems for autonomous landing have been a very active field of research. A typical landing operation comprises 3 phases: during navigation phase, the UAV returns from the mission and reaches the Final Approach Fix; during guidance phase, UAV is controlled by GPS or other means to the Guidance Window; finally, during landing, the operator in the GCS supervises the procedure by means of the camera image from the laser sensor and data from the flight control. LECTURE 5 Describe a typical UAS guidance and control architecture (see slide 5). Which are the implications of flight autonomy on the required piloting skills, and on the tolerance to communication latency and communication issues? An unmanned aircraft system is a dynamic system that basically reacts to control inputs and disturbances. In the figure 1.3, two different possibilities of creating a UAS are represented. In the first case, servo- commands (or control inputs), together with the disturbances (like wind) arrives directly to the UA, providing information about velocity, position and attitude. Such information is then processed by the Navigation System. The pilot must be skilled and the delay or latency of the command is really short. In the second configuration, the pilot does not directly give control inputs, rather higher level information such as airspeed to be reached by the UAV. The autopilot evaluates the servo-commands needed for unmanned aircraft. The stability of the aircraft is therefore provided by the autopilot. In many UAS applications we find a guidance algorithm providing higher level commands to the autopilot. Autopilot provides information about the desired trajectory to a path following block, whose role is to obtain the higher level commands needed by the autopilot. It is also possible to insert a higher level of processing inside the UAS, using a path manager block that can evaluate the desired path by simply giving it some desired points; path manager elaborates the desired trajectory that the path following must follow. Then we can insert another higher level software, called path planner which can evaluate the points of trajectory and optimize the choice. LECTURE 6 Describe the basic operation of an autopilot: which are inputs and outputs (see slide 4)? The autopilot system works as a MIMO system that elaborates multiple inputs in order to produce multiple outputs. Basic autopilot inputs are attitude and directional gyros, turn coordinator, attitude control, etc.; these are all sensing units of position, velocity and attitude that sense the which is the phase angle of relative position. For longitudinal guidance we have constant altitude command, while for lateral guidance we have a vector field guidance as in the straight-line path. The control objective is to drive d(t) to the orbit radius p (rho) and to drive the course angle X(t) to X° in the presence of wind. LECTURE 8 Describe waypoint switching principles (slide 6). A waypoint path is defined as an ordered sequence of waypoints. The simplest path is a sequence of straight segments and we mainly use two methods to switch from one segment to the following one: - the first is called b-ball around method, and it consists on the avoidance by the UAV of the imaginary ball around the relative waypoint; - The second method is called half plane through waypoint. The waypoint switching is performed when half of the plane is through the waypoint. This is considered a robust strategy and a very commonly used method to perform waypoint switching in Unmanned Aircraft Systems. How can we smooth transition between path segments (slide 9)? Waypoint following is extremely simple: the UAV reaches a waypoint before transitioning to the next straight-line path. As a drawback, this process results in a non smooth or unbalanced transition between straight-line segments; thus we insert fillets to connect straight-line paths and smooth transitions between segments, to make the process more realistic. In most cases the path can be smoothed by adding a fillet before the waypoint. The turning radius must be compatible with aircraft capabilities; a larger radius means smoother manuevers but also larger distances from the waypoint. What are Dubins paths? Why are they useful? In geometry, the term Dubins path typically refers to the shortest curve that connects two points in the Euclidean plane (x;y); in a kinematic model, the shortest path will be obtained by joining circular arcs of maximum curvature and straight lines. Dubins paths are a particular kind of paths that are widely used in path generation; they consist of different pre-defined configurations, in terms of velocity and position. A certain configuration is always individuated by position and course angle and the objective is to transition from one configuration to another. We are interested in the shortest time trajectory to pass from the initial to the end configuration, assuming that the vehicle has a constant groundspeed, deriving from the fact that shortest time also results in shortest length, constant altitude (two dimensional space) and turning rate limits of a certain vehicle, for example the maximum turn radius achievable corresponding to a minimun turn radius). For vehicles with kinematics, Dubins paths are given by 3 equations: If the initial position of the airplane and the related speed vector is located in another point, then we could have different combination of path following. In all cases the Dubin Path chosen will be the shortest of four cases, in general this choice is computed numerically by the system. Dubins path can be extended to the three dimensional space, which is more complicated due to the altitude component that must be taken into account. According to altitude difference between start and end configurations, length of the 2D Dubins path and aircraft path limit we can have 3 different configurations: -in a low altitude path, the 3D path is obtained by modifying the 2D path with the necessary flight path angle that must be lower that the maximum admitted by the airplane; - in a medium altitude path, the 3D path is obtained adding an intermediate arc to the modified 2D path; -in an high altitude path, the 3D path is obtained by adding spiral turns to the modified 2D path. LECTURE 9 Which are the basic path planning problems (slide 4)? The Highest level of autonomy defines waypoints (or configurations, that is, position/velocity couples) based on mission scope and constraints. Path planning has two approaches: -deliberative approach, based on global world knowledge, which requires a good map of terrain, obstacles. It can be too computationally intense for dynamic environments and is usually executed before the mission. -reactive approach, based on what sensors detect on immediate horizon. It can respond to dynamic environments and is not used for the entire mission. For what concerns problems related to path planning we can say that there are two main challenges: -concerning point to point, the objective is to pIan a waypoint path from one point to another through an obstacle field; -regarding coverage, the objective is to plan a waypoint path so that the UAV covers all of the area in a certain region. In reality many different approaches exist to solve the problem of building a graph and navigating through it; one of the most important examples is the Rapidly exploring Random Trees (RRT) approach. Describe the principles of RRT algorithm. RRT algorithm is able to calculate the better path considering obstacles and possible collisions for the UAS. It is an exploration algorithm that randomly, but uniformly explores search space; it can accommodate vehicles with complicated dynamics and obstacles represented in a terrain map. Map can be queried to detect possible collisions and RRTs can be used to generate a single feasible path or a tree with many feasible paths that can be searched to determine the best one. If algorithm runs long enough, the optimal path through the terrain will be found. In order to have the algorithm initialized we must have a start node, an end node and a terrain/obstacle map. We randomly select a new configuration p in workspace and select a new configuration v1, with a fixed distance D from start point along line connecting ps and p. Then we insert v1 into tree and check for new segment for collisions and if collision occurs, we delete the segment. After that we once again generate random configuration p in workspace and search tree to find node closest to ps. From node closest to p, we move with distance D along line connecting node and p to establish new node v2, which is inserted into tree and we check new segment for collisions and if collision occurs, we delete segment. We must continue until node is generated that is within distance D from the end node. At this point, algorithm is terminated or we search for additional feasible paths considering that path can then be smoothed taking aircraft dynamics into account. LECTURE 10 Describe quadrotor control principles. In the field of multi-rotors, quadrotors are the most standard configuration. A quadrotor consists of 4 individual rotors, 2 rotating clockwise and 2 counter-clockwise attached to a rigid cross airframe. Quadrotor control is achieved by control of the thrust generated by each rotor. The total thrust is the sum of thrusts generated by each rotor and it is used to control vertical thrust; pitch and roll are obtained by differential thrust along pitch and roll axes; heading rate control is achieved by differential control of the counter-clockwise rotors compared to the clockwise rotors, interference (ASI). Geostationary SATCOM satellites are provided with transponders, which are high gain antennas that provide ground coverage over a designated spot on the Earth's surface. The resulting pattern is called a spot beam. The higher the transponder gain, the smaller the ground coverage. The transponders are usually steerable, and so the geographic coverage selection is driven by the customer demand. SATCOM satellites can be owned either by governments or by companies. A minimum elevation angle is required for SATCOM. What do we mean by "remote split operations"? BLOS communications require relay nodes to retransmit the signal. Satellites are the most common relay nodes and they have the advantage of being able to transmit the signal even globally in remote split operations. This type of operations is mainly used in military applications and they allow fast communications from off-site pilots to individual unmanned vehicles; as a result, pilot can fly aircraft that are thousands of miles away. What do we mean by "hub and spoke operations"? In the field of ground control stations (GCS) for tactical, MALE and HALE systems, we can talk about a particular type of operations commonly used by airlines, which are hub and spoke operations: they use a launch and recovery hub site to provide the UA service capabilities in multiple regions; the hub is the LRE and the spoke is the FGCS. What do we mean by GCS/Unmanned Aircraft interoperability? A single GCS that can fly multiple UA types can give significant cost and operation advantages. Thus, there has been an increasing effort in developing an open air-to-ground interface standard, so as to decouple UA and GCS developments. This effort has resulted in the NATO STANAG 4586. It was developed to provide a level of UAV interoperability across coalition forces to allow the ability for quickly tasking available assets, to mutually control these vehicles and their payloads and to rapidly disseminate tactical information to the collective force. STANAG 4586 consists of the core UAV control system (CUCS) and a vehicle specific module (VSM). The CUCS includes processing, mission planning tools, and user interfaces. The CUCS remains substantially unchanged for flying different UA, though it can be tailored. The VSM is a module that is specific to the UA. The VSM has a standardized interface to the CUCS, and it can be implemented on the ground or on the UA. STANAG 4586 is an evolving standard with frequent revisions. There are multiple levels of interoperability defined within the standard, regarding receipt of payload data, payload control, UA flight-path control (exclusive of launch and recovery), launch and recovery control. LECTURE 12 Which are the basic frequencies used for UAS communications? Describe advantages and drawbacks of high and low frequencies. Frequencies in the range 3 Hz to 3 GHz are generally considered to be the true radio frequencies as they are refracted in the lower atmosphere to curve to some degree around the earth’s circumference, increasing the range of line-of-sight operation. Frequencies above this range, between 3 and 300 GHz, are not refracted and therefore operate only line-of-sight. It is necessary to transmit high rates of data, especially from imaging-sensor payloads, from the aircraft to its control station or other receiving station. Only the higher radio frequencies are capable of doing that and, unfortunately, these depend progressively towards requiring a direct and uninterrupted LOS between the transmitting and receiving antennas. There is therefore a compromise to be made when selecting an operating frequency: a lower frequencies offer better and more reliable propagation but have reduced data-rate ability, and the higher frequencies are capable of carrying high data rates, but requires increasingly direct LOS and generally higher power to propagate the signal. Frequencies in the range 1–3 GHz are a desirable compromise in most circumstances, but due to increasing demand by domestic services (ex. television broadcasting) for the use of frequencies in the VHF, UHF ranges, the frequency allocation agencies are requiring that communication systems use increasingly higher frequencies into the microwave band of 5 GHz or above. Indeed, frequency allocation is considered as a critical point for the future of UAS, and frequency conflicts could be one of the limiting factors hindering UAS diffusion in civil scenarios. Which are the factors that may produce loss of communications during operations (see slide 4)? Communication systems provide the means to distribute data among system elements and to external entities. UAS usually use radio-frequency (RF) communications systems to transmit data wirelessly. These systems are configured for either direct line-of-sight or indirect beyond line-of- sight communications such as SATCOM or airborne relay. The maintenance of the communications is of paramount importance in UAS operations, because without the ability to communicate, the system is uncontrollable and useless. Loss of communication during operations may result from: 1)a failure of all or part of the system due to lack of reliability; 2)loss of LOS due to geographic features blocking the signals; 3)weakening of received power due to the distance from the UAV to the control station becoming too great; 4)intentional or inadvertent disturbance/jamming of the signals. The specifications for communications performance include two fundamental parameters, which are data rate, that is the amount of data transferred per second by a communications channel and is measured in bytes per second (Bps) and bandwidth, which is the interval of frequencies the the amplitude spectrum of the communication channel assumes it significant values. The larger the bandwidth, the faster the data transmission speed. Which are the typical antennas used for UAS communications? Antennas represent the transition between transmission lines and free space propagation of the electromagnetic field. Antennas transmit and receive RF energy and provide directivity. Their size and configuration are limited by constraints and affect performance and beam pattern characteristics. Highly directional antennas need to be pointed with mechanical and/or electronic steering: onboard pointing is similar to EO gimbal pointing, but the field-of-regard requirements can be different. Typical antennas used for UAS communications are the following: -the quarter wave dipole antenna is used at all frequency bands, provides good performance, omnidirectional radiation pattern and simple construction; -the Yagi-Uda antenna has a reflector in addition to a number of directors. Gain and directivity of this antenna enable better reception due to better levels of signal to noise ratio and by reducing interference levels by only picking up signals from a given direction; -the parabolic reflector antenna has high gain and frequency and it is pointable. All reflected rays have the same distance from a focus point to a flat opening aperture plan; -lens antenna and phased-array antenna. Describe the main communication system types (slides 21-25). The main communication systems types are command and control, payload link, air-to-air communications, voice relay, transponders. The first two categories are the most important in intelligence, surveillance and reconnaissance missions (ISR). Air-to-air communications enable greater radio coverage by airborne relay or they can be used in cooperative missions employing several UAVs operating together (in these cases, networked radio links are used). Voice relay and transponders are important in civil missions where the UAS has to share the same airspace of manned aircraft. VOICE RELAY| A UAS operating in the same airspace as manned aircraft is often required to behave like one of them. Manned aircraft can have voice communications with the ATC system. There are two methods for satisfying operator-ATC communications: 1)voice communications between the UA operator in the ground control station and the ATC can be relayed via the unmanned aircraft or 2)the ground control station can have direct communications with ATC through ground radios, though this has range limitations because of terrain blockage and the Earth’s curvature. For long-range UAS controlled through SATCOM, ATC voice communications through satellite can be implemented. If multiple satellite hops are required, then the latency can cause effective communication difficulties. C2| UA command and control (C2) data Iinks provide the operator with the ability to direct the flight and understand the unmanned aircraft state. This link is critical when the aircraft is incapable of landing by itself. C2 Iink must be secured against jamming and unauthorized use, especially for military operations, and so anti-jamming and encryption capabilities are recommended. Unintentional RF interference is also a risk so a backup C2 link provides redundancy. C2 is generally low bandwidth. The low bandwidth requirement is compatible with frequencies in the VHF and UHF spectra for line of sight operations. LEO (Low Earth Orbit) satellite constellations can also provide BLOS C2 capabilities. Considering C2 links, communication hand-off between two GCS is a common operation. For example, one GCS is responsible for launch and recovery (LOS), and another one is responsible for mission operations (LOS or BLOS). There are two approaches to hand the C2 link from one GCS to the other: make before break, where both GCS have positive C2 RF communications established before the first one relinquishes command authority, and it is the safest approach but require two C2 links onboard the aircraft; break before make, where the first GCS shuts down the C2 link, and then the second establishes C2 link with the unmanned aircraft. This procedure has the risk that the unmanned aircraft is momentarily out of control. Describe different scenarios for Sense and Avoid systems, based on the physical location of information sources and processing/decision making center (slides 14- 18). The three fundamental tasks of an SAA system can be implemented in different ways, giving rise to several architectural and technical solutions. These solutions can then be classified using different taxonomies. A general approach for classification that involves all three parts is based on the physical location of information sources and processing/decision making centers. These elements can be based onboard the UA, or they can be located on the ground. Some different scenarios for SAA systems. Considerations. Exploiting off-board systems simplifies technological challenges, especially regarding sensing tasks and compatibility with aircraft budgets: ground-based architectures can work with little or no modifications of existing UAS designs. Possible drawbacks are the limited coverage of ground systems and the dependency on C2 link. As a consequence, off-board SAA is usually considered to be a local and/or near-term solution for meeting requirements to enable integration in limited operational areas, while onboard SAA is the definite key goal to be accomplished for future systems. Describe the general classification for Sense and Avoid systems (slide 20). SAA systems can be classified according to 4 different macro-areas: sensing, conflict detection, avoidance, architecture. Which sensing technologies can be used for Sense and Avoid? Which is the difference between cooperative and non-cooperative sense and avoid? Sensing technologies used for SAA can be cooperative or non-cooperative; while the first require intruder aircraft to be appropriately equipped and exchange of information between aircraft, the second ones do not require the participation of the intruder aircraft and detect collision threats without external support. Non-cooperative techniques can then be active (requiring electronic interrogation) or passive (no need of electronic interrogation). Cooperative techniques include ADS-B (Automatic Dependent Surveillance-Broadcast) and TCAS interrogator/transponder; non- cooperative techniques include Electro Optic (EO) and acoustic (both passive), primary radar and LIDAR (both active). What are distance and time to closest point of approach (slide 39)? In conflict detection there are some general and important variables for 3D conflicts detection: the distance at the closest point of approach and the time to the closest point of approach. Both involve position and velocity and because of this are based on Tracking Algorithms. The following equation define the distance at the closest point of approach considering the side approaching rappresentation. Another very important topic reguards the conflicts conditions that are verified only if: time to closest point of approach Which are the principles of conflict detection based only on angles and angular rates (slide 43)? Indeed, Conflict detection can be addressed without considering the availability of range information, at least at a certain level of accuracy. This is what happens for human pilots, who are able to routinely solve the see-and-avoid problem in absence of accurate range information. A future collision condition can be foreseen based on angular-only information that is if the line of sight rate of an obstacle (North-East-Down axes) is zero (there must be also some evidence that range rate is negative, such as increasing obstacle dimensions in pixels).