From AI-powered missiles to drone swarms, how Iran is reshaping its aerospace sector with artificial intelligence

By Yousef Ramazani

Artificial intelligence has become an increasingly important component of Iran’s aerospace industry, where domestic experts and technologists have developed and localized technologies that integrate machine learning, image processing, and autonomous systems into applications ranging from satellite observation and coordinated drone flights to advanced missile guidance and integrated air-defense command and control.

It comes as the world stands on the threshold of a new technological revolution in which artificial intelligence is no longer a peripheral tool but increasingly functions as the intelligent core of both defense and commercial systems.

The integration of artificial intelligence into modern industries is reshaping traditional concepts of technological power, enhancing the deterrence, operational efficiency, and autonomous capabilities of states.

Iranian technologists and researchers, recognizing the rapidly changing global technological landscape, have focused on combining artificial intelligence with aerospace technologies and developing domestic capabilities in advanced machine learning, computer vision, and image-processing algorithms.

Achievements include defense systems designed to analyze threats and support rapid decision-making, as well as coordinated drone systems capable of operating collectively and autonomously.

The integration of artificial intelligence with aerospace technologies is no longer confined to the military sphere. It is also emerging as a driver of transformation across strategic and commercial sectors, from satellite-data analysis and precision agriculture to navigation, environmental monitoring, and defense systems.

Iran has increasingly moved beyond the experimental and conceptual stages toward the practical deployment of these technologies. Artificial intelligence is consequently playing a growing role in improving productivity, operational accuracy, technological self-reliance, and the country’s broader aerospace capabilities.

Iran's artificial intelligence portfolio is nothing short of impressivehttps://t.co/LPNyKs4raJ

— Press TV 🔻 (@PressTV) February 26, 2025

Integrated command and control: Great Prophet and Payambar-e Azam systems

One of the most important applications of intelligent computing in aerospace and air defense is command and control. An air-defense network must process information from multiple sources, including surveillance radars, passive sensors, electro-optical systems, and observation posts.

Each sensor provides only a partial view of the environment, while differing in accuracy, update rates, coverage, and technical limitations.

The purpose of an integrated command-and-control system is to transform these disparate observations into a common operational picture, enabling commanders to assess threats and make informed decisions rapidly.

Iran’s Payambar-e Azam integrated air-defense command-and-control system, together with tactical systems such as Fakour, represents an indigenous software and communications architecture designed to connect sensors, command centers, and defensive units.

Descriptions of the system include capabilities such as target tracking, threat assessment, decision support for missile systems, weapon allocation, and the prediction of potential enemy movements.

The system is intended to provide commanders with a consolidated picture of the battlefield, allowing them to evaluate incoming information and make timely operational decisions.

According to insiders, Payambar-e Azam incorporates artificial intelligence to support target tracking and prediction, accelerate decision-making, and process and integrate large volumes of data received from multiple sources. These capabilities are intended to facilitate complex analysis, identify potential threats, and improve the overall efficiency of air-defense operations.

A modern air-defense command-and-control system does not necessarily “think” like a human. Instead, its computational functions typically include several interconnected processes.

Sensor fusion combines radar, electro-optical, signals-intelligence, and other observations to create a more complete picture of individual objects. Track management estimates the location, speed, and trajectory of aircraft, missiles, or UAVs and predicts their likely future positions.

Classification helps distinguish different types of objects based on their observed characteristics, while threat assessment determines which tracks require the greatest attention. Resource management then helps match available defensive systems and weapons to identified threats.

Finally, the human-machine interface presents the processed information to commanders and operators, allowing human decision-makers to interpret the overall situation and determine an appropriate response.

Sahm: Applying machine learning to satellite imagery

The Sahm intelligent satellite-observation system, developed by the Iranian company Tiznegar, represents one of the more documented applications of artificial intelligence in Iran’s aerospace sector.

The system consists of two main components. The first is a satellite-image processing engine. Since 2019, the company’s research and development team has been designing and developing multispectral satellite-image processing algorithms in Python to detect and analyze environmental phenomena.

The algorithms and mathematical models are developed using ground-based data, as well as larger-scale maps and reports. These datasets provide reference material for algorithm testing and validation, as well as for machine-learning training.

The basic principle behind Sahm is relatively straightforward. Earth-observation satellites record reflected or emitted electromagnetic radiation across several spectral bands. Vegetation, water, soil, buildings, smoke, and other materials interact differently with different wavelengths. Computers can therefore analyze these spectral differences to identify land-cover types and detect environmental changes.

Machine learning can further improve this process by training algorithms on examples for which the correct classification is already known. Ground observations, cadastral information, and previously interpreted satellite imagery can serve as training and validation data.

Once trained, a model can classify new satellite images and generate geographic layers identifying phenomena such as agricultural land, vegetation stress, water bodies, fires, and other environmental changes.

For example, an algorithm designed to identify cultivated areas can use data collected through land surveys and cadastral records to establish quantitative criteria for evaluating and analyzing satellite imagery. Once the image-processing algorithm produces its results, the information can be linked to descriptive databases and converted into location-specific reports.

The second component is a Web GIS platform. This platform creates and maintains databases containing spatial and descriptive information generated from the satellite-image processing algorithms, allowing the results to be analyzed, reported, and displayed geographically.

Sahm is therefore more than a conventional image-processing program. Its architecture combines an image-processing engine with a GIS environment, enabling processed satellite imagery to be connected with geographic and descriptive databases and presented through maps, analytical outputs, and reports.

Descriptions of the system have identified applications in agriculture, flood monitoring, fire detection, water-resource management, and the monitoring of atmospheric pollutants. The system illustrates how artificial intelligence can transform satellite imagery from raw visual data into structured geographic information for environmental monitoring and resource management.

Coordinated drone flights: Swarm intelligence in the sky

Another major area of development is coordinated multi-UAV flight, in which multiple drones operate simultaneously under a shared control and planning system.

According to insider accounts, a domestic technology company, after obtaining the necessary permits from the Civil Aviation Organization and other relevant authorities, developed localized technologies for intelligent flight systems and the collective control of drones.

The system was demonstrated for the first time in Qazvin Province, where dozens of drones performed coordinated flights, formations, and aerial displays using a locally developed planning system. The demonstration required precise synchronization between the aircraft, as well as advanced navigation, communication, and flight-management technologies.

Coordinating a group of drones is considerably more complex than programming a single UAV to follow a predetermined route. Each aircraft must continuously maintain its own position, altitude, speed, and orientation while also maintaining an appropriate relationship with the other vehicles in the formation.

A functional multi-UAV system therefore involves several interconnected capabilities. Localization allows each aircraft to determine its position; communication enables drones to exchange information about their status and surroundings; and formation control determines their relative positions.

Collision avoidance maintains safe separation, while trajectory planning establishes appropriate flight paths. Synchronization ensures that coordinated maneuvers occur within the required timing tolerances, while fault handling allows the formation to respond to problems such as a lost aircraft or communications interruption.

During the Qazvin demonstration, drones equipped with autonomous flight algorithms and advanced communications systems performed synchronized flights, coordinated formation changes, geometric patterns, and planned individual and group maneuvers.

The underlying technology draws on concepts such as swarm intelligence and multi-agent control, in which individual aircraft follow distributed rules while coordinating their behavior with other members of the group. Artificial intelligence can support this process by helping automate the coordination, synchronization, and adaptation of multiple aircraft.

The same fundamental technology has potential civilian applications, including synchronized aerial displays, infrastructure inspection, surveying, search and rescue, disaster response, and environmental monitoring.

The significance of multi-UAV technology therefore extends beyond simply enabling many drones to fly together. The more important transition is toward distributed autonomy: a system in which multiple aircraft can coordinate their behavior while continuously responding to information from their own sensors and from the wider network.

Qassem Basir: Advanced missile guidance with AI-assisted targeting

The Qassem Basir missile, capable of reaching targets more than 1,200 kilometers away, represents one of Iran’s most significant recent developments in missile technology.

Iran publicly unveiled the solid-fuel ballistic missile in May 2025, with officials describing improvements in guidance, maneuverability, and its ability to operate without reliance on GPS.

The missile was tested on April 17, 2025. Iranian officials say that it can distinguish a designated target among multiple objects and achieve accuracy at approximately the meter level.

The missile incorporates an advanced electro-optical or imaging seeker designed to support target recognition and terminal guidance. Such a system can provide the missile with visual information about its surroundings during the final stage of flight, supplementing its internal navigation system.

Expert descriptions have associated artificial intelligence with two principal functions. The first is navigation, which means machine-learning techniques could potentially be used to analyze sensor data, compensate for navigation errors, and improve estimates of the missile’s position and trajectory under changing conditions.

The second is machine vision, in which image-processing algorithms can assist an electro-optical seeker in identifying and tracking a designated object during the terminal phase.

The use of electro-optical terminal guidance is technologically significant because imaging sensors provide information about the target environment that differs fundamentally from the data available through inertial navigation alone.

This can allow a guidance system to update its understanding of the target during the final phase of flight without depending exclusively on satellite-based positioning.

The missile is designed to operate in environments involving electronic countermeasures and has been successfully tested under demanding conditions.

AI-assisted aerial mapping: intelligent reconnaissance UAV

Another example of AI integration in Iranian aerospace is an unmanned aerial vehicle capable of being remotely controlled and used to perform mapping operations and identify specific objects in images in most fields, including construction, agriculture, and other applications.

The UAV is equipped with machine vision, including an imaging camera as well as image processing and machine learning online at the ground station.

In the field of agriculture, it can be used to identify plant species, the color of plants and tree leaves, and identify images in the form of textures at high altitudes.

In addition to high accuracy and speed, this drone provides the ability to pass through dangerous and difficult-to-pass areas at any time interval and enables the use of non-metric cameras instead of metric cameras, which is a particularly interesting technical feature.

The system consists of three hardware parts: navigation, data acquisition, and ground station, as well as a software part. The UAV software includes navigation control, flight planning, preparation and processing of images received from the ground, and analysis of image content by artificial intelligence.

In this reconnaissance UAV, artificial intelligence plays a central role in machine vision and automatic image analysis. The UAV system uses machine learning algorithms to process image data in real time at the ground station to automatically identify objects, patterns, or specific features such as plant type, leaf color, or ground texture.

This technology enables the drone to make intelligent decisions about flight paths or important points for mapping, providing precise spatial and image data without the need for direct human control.

If the machine-learning processing occurs at the ground station, the system is better described as an AI-assisted aerial mapping platform rather than an entirely autonomous AI aircraft, which is actually more interesting because it illustrates a practical architecture: airborne sensing, communications, ground-based computing, machine learning, and GIS.

Broader significance of Iran's aerospace AI developments

Taken together, these examples demonstrate something more significant than Iran’s growing use of artificial intelligence. They point to the emergence of a broader technological architecture in which sensors, communications networks, computing systems, and automated decision-support tools are becoming increasingly interconnected.

Satellite imagery can be transformed into geographic intelligence through machine learning, while UAVs can use computer vision to convert aerial photographs into maps and analytical information.

Multiple drones can coordinate their movements through distributed control systems, and air-defense networks can integrate information from numerous sensors into a common operational picture. Advanced missiles can likewise combine inertial navigation with electro-optical sensing and terminal guidance.

The significance of these developments lies less in the idea that Iran has created autonomous machines capable of replacing human judgment and more in the gradual integration of artificial intelligence into sensing, interpretation, coordination, and decision support.

The localization of satellite-image processing technologies through systems such as Sahm, together with the practical demonstration of coordinated drone flights, illustrates how domestic research and engineering capabilities can be used to develop specialized technologies and reduce reliance on foreign systems.

Artificial intelligence is consequently becoming an increasingly important strategic technology for Iran’s aerospace sector. Its applications extend from monitoring the Earth and managing natural resources to navigation, autonomous systems, and air defense, reflecting a broader global shift toward the integration of intelligent computing with aerospace technologies.

Press TV’s website can also be accessed at the following alternate addresses:

www.presstv.ir