Novel scheme by IIT Bombay researchers to control drones can enable complex formation flying using only camera data, without GPS or inter-drone communication.
A new control scheme developed by Professor Dwaipayan Mukherjee and research scholar Chinmay Garanayak at the Indian Institute of Technology (IIT) Bombay enables unmanned aerial vehicles (UAVs) to fly in coordinated swarms without relying on GPS, inter-drone communication, or centralised control systems. The method uses bearing-only measurements obtained through onboard cameras to determine relative positions and maintain formation.
The researchers applied the scheme to Vertical Take-Off and Landing (VTOL) UAVs, which can lift off without a runway and hover mid-air. These drones are suitable for operations in confined spaces such as surveillance and monitoring. “Autonomy in a swarm is a critical task,” Mr. Mukherjee said. “This means that vehicles in a swarm should be able to decide their ‘actions’ based on variables they can measure with their on-board sensors, instead of having to rely on some global information being fed to them or some human/centralised computer deciding what their action ought to be. This is where our paradigm differs from usual ones,” he added.
The proposed ‘bearing-only’ control scheme allows each drone to use its onboard camera to observe its immediate neighbours and calculate bearing information. “In bearing-only control, the goal is to achieve formation control using only interagent bearing measurements,” Mr. Garanayak said. The system does not require GPS or communication with other drones or a central computer.
Camera-based measurements are less prone to noise than conventional distance sensors, simplifying the drone’s sensor system and reducing battery requirements and overall weight. The scheme is designed to work in areas where GPS is unavailable, or communication could be jammed, making it suitable for stealth-mode operations such as covert military missions.
VTOL drones are underactuated systems, which means they have six degrees of freedom but fewer directly controllable movements. While they can move vertically and rotate around three axes, lateral and forward-backward movements must be indirectly controlled. “Many of the results in the literature do not address the underactuated dynamics of VTOL vehicles and only focus on the kinematic model. This motivated us to consider the fully underactuated model of the VTOL UAV and explore its applicability to formation control,” Mr. Mukherjee said.
Underactuated systems require dynamic models that include position, orientation, velocities, forces, torques, and inertia. Previous attempts to apply bearing-only control to such models often fail due to instability or breakdowns in certain conditions. Mr. Mukherjee and Mr. Garanayak developed a control mechanism that ensures convergence and maintenance of the desired formation, even when drones start from imperfect positions. They have provided rigorous mathematical proof to support the reliability of the system.
Their work addresses two operational scenarios. In the first, drones maintain formation at constant velocity using bearing and bearing-rate data. In the second, where formation and velocity vary over time, drones incorporate their own velocity measurements in addition to bearing data. The system can handle arbitrary time-varying configurations, allowing drones to navigate narrow passages, reconfigure into single-line formations, and adapt to changing mission requirements.
The researchers plan to test the control scheme experimentally, using a drone swarm. On the future roadmap, they aim to address collision avoidance with theoretical guarantees. “Most existing algorithms rely on ad hoc collision avoidance schemes that do not come with any theoretical guarantees. Collision avoidance with objects in the environment and among drones is a challenge we are trying to tackle at a theoretical level,” Mr. Mukherjee said.
Published – November 02, 2025 07:44 pm IST
