A research group from a Spaniard university has designed a new system based on optimal control technology for Air Traffic Management (ATM). The new approach is mainly characterized by its adaptability capacity in situations where there are more than two aircrafts. Potential collaborations with international/national research centres and companies interested in applying the technology are sought.
New and innovative aspects
The developed technology for being applied to aircrafts in order to perform an Air Traffic Management (ATM) is cost-effective and robust against changes in the environment or in physic structure of the platform. One of the pillars of the new technology is the lack of mathematical models for the performance of control and, instead, it uses the experience gained through interaction with the environment, the platform and communication with other aircrafts.
Main advantages of its use
-Low cost, robust against changes, air traffic management optimally and real-time.
-Easy transfer of technology due to the maturity achieved with the new approach in other areas.
The Space Research Group (SRG), specialised in different areas related to Aerospace, including Optimal Control, designs and builds instrumentation to be used on-board satellites, spacecrafts and unmanned aerial vehicles.
The technology offered by this group, is able to perform Air Traffic Management (ATM) on aircrafts of an optimal way according to specific optimization criteria (e.g. minimum time).
The basis of the research carried out by the SRG Group is to consider two areas of interest. A first area called critical, and a second-limits to certain aircraft. The first zone has a volume greater than the second and will be where the aircrafts arrive first, according to their path. The new technology works as follows way: The first aircraft that comes into the critical zone will not modify its trajectory. However, for the second and successive will apply the new technology to alter its trajectory of an optimally and minimally way, to avoid any possible collision in the second zone.
It should be noted that if the trajectory of an aircraft has to be changed, this will be done in the critical region. Also, as stated above, the trajectory of the first aircraft, coming into the critical zone will not change. Therefore, this aircraft will not apply any control. However, for the second and successive aircrafts, an optimal control in real-time will be applied.
The alteration of the trajectory of an aircraft implies a communication with other aircrafts in order to know their particular trajectories, and in this way, avoiding possible collisions. In order to modify a trajectory, first it is necessary to set a reference location inside critical region. This location will be an obstacle region, for the second aircraft and thus, it will have to dodge it optimally. On the contrary, for the first aircraft, it will be a free region. However, two obstacle regions are considered, for the third aircraft: the same than the second plus the region that second creates for avoiding the first one. This reasoning can be applied to the fourth aircraft and subsequent.
The aircrafts are characterized by so-called state variables and the procedure carried out by the new technology is to act on the aircraft through control actions in order to modify its current state. Therefore, the set of state variables and their values range constitute the so-called state space which will be the scope of the new system. Due to the dynamics of the aircrafts is continuous, it is necessary to consider a discretization of the state space when finding optimal controllers. Thus, the concept of cell arises, and the obstacle regions mentioned above are composed of several cells.
Applying the new technology that the SRG Group has developed lets us have an available system, which is very robust against changes in the environment or in physic structure of the aircraft. In order to achieve this robustness, it is necessary to perform a learning that takes into account of an implicit way these possible changes and is able to know the dynamics of the aircraft in real-time. From this learning, the system shall perform optimal navigation planning for changing the original path and in this way, avoiding any kind of collision. Also, the new approach is a closed-loop solution and learning is focused on reaching a specific goal (dodging the reference location inside critical region) starting from any origin state (when coming into critical region), according to an optimization criterion. The origin and goal states are characteristic states from a point of a dynamical of the aircraft view.
In this way, the aircraft learns the dynamics and kinematics from the experience in real-time without any kind of explicit mathematical models. Thus, the implementation of the new technology consists of learning the dynamics of the aircraft and performing optimal control of a concurrent way. The great advantage of the concurrency is to adapt the controller while it is being applied and, thus, achieving a behaviour that comes close to optimum.