Russian defense Ministry conducted a test of a unique software which gives unmanned aerial vehicles, artificial intelligence.
Joint development the Main research and testing robotics centre of the Ministry of defence (GNIIT RT) and scientific-production company “network-centric platform” (TS SP) turns private drones in self-organizing group, capable not only to carry out the joint flight, but also to conduct patrols of facilities, as well as to find and destroy ground and air targets of the enemy.
— Testing and computational simulation of different tasks took place from March to August of this year. In particular, we have successfully fulfilled the control of unmanned aerial vehicles in the implementation of the joint flight with the transmission of intelligence information from various angles and directions, — told “Izvestia” head of NPK SP Alexander Mochalkin.
Is, in particular, helped to create three-dimensional computer model of the tactical situation in the area of application of the devices. Earlier last year as part of the civil program of precision agriculture, we have worked a similar situation in practice, in the Rostov region. Then we combined the “flock” of three drone quadcopter for accurate terrain mapping.
According to Alexander Mochalkin, now NPK SP is able to combine in a “pack” of up to six drones at a range of their regular channels.
The principle of operation of the innovative system quite simple. The drones are equipped with special software and a small computing unit that is able to calculate the future behavior on the basis of the received command and data the onboard sensors of the drones.
To control the “flock” base station — staff control unmanned aerial vehicles with installed software package that turns the operator into a set of algorithms that are transmitted and pack.
As told “Izvestia” in the defense Ministry, currently GNIIT RT and NPK SP moved on to the next stage of the work, in which drone “flock” will teach together with air defense to intercept the rapidly moving air targets — aircraft, helicopters, cruise missiles and enemy drones. Also “flock” will learn to make and impact, in particular, to break the echeloned air defense system of the enemy.
— The UAV should operate independently, choosing the tactics of the attack on the target, depending on the prevailing tactical situation, — told “Izvestia” a source in defense Ministry familiar with the process. — In this part drones should play the role of “victims”, the disturbing forces and means of defense for themselves and others — to directly hit the target. In this work, the planned installation of new software in the control equipment of combat UAVs, as well as fitting them with electronic control units of the company “network-centric platform”.
Professor of the Academy of military Sciences Vadim Kozyulin told that works on creation of integrated systems of management and coordination of unmanned aerial vehicles in the armies of developed countries is considered one of the top priorities.
— Currently in the armed forces of the United States, France, Britain and other NATO countries have a large number of unmanned air and ground systems, but the management in most cases is either pre-stored program or by the operator from the ground, — said Vadim Kozyulin. — At violation of the communication channel all of these robots immediately cease to operate. That is why questions of creation of mathematical models of artificial intelligence based on the principles of self-organization for the development of robotic systems are today a priority.
The expert also noted that the software coordination and targeting of unmanned aerial vehicles is implemented on the basis of helicopter AH-64E Apache Gurdian, recently adopted by the U.S. army.
In our design incorporates a completely new innovative mathematical apparatus based on multi-agent technology, — explained the “Izvestia” Alexander Mochalkin. — A control system for unmanned aerial vehicles based on their representation as software agents. In fact, it is an independent specialized computer software in addition to the standard autopilot of the aircraft.
She is responsible for the coordination of actions and adoption of decisions by the UAV during the flight assignment. For example, they should automatically classify the target on the previously laid electronic portraits, organize activities for the formation of desired spatial and temporal structure of movement in range of the enemy’s air defense systems.
