AEROCADASTRA
A fixed-wing mothership, two autonomous child drones, and a shared coordination layer with three independent live video feeds converging on one ground station.
The mothership is an original, team-designed fixed-wing airframe intended to carry and coordinate two child quadcopters. Its performance remains subject to bench and flight validation.
The current video baseline uses an MS MINI camera and DakeFPV video transmitter. Imagery and AI perception are intended to support detection and mapping; integrated performance is not yet validated.
GNSS provides positioning for navigation and geo-referencing. RTK/PPK precision is a future upgrade, and cadastral-grade accuracy must be demonstrated through measured validation before it is claimed.
The two child quadcopters are designed to work with the carrier through shared identity, telemetry, deployment and video interfaces. The three-feed system remains a core integration target.
The target architecture brings fleet telemetry and three independent video feeds to one ground station, with ROS 2 software, AI perception and geospatial processing. End-to-end integration still requires validation.
Mapping the land, faster
AeroCadastra is developing a modular autonomous aerial fleet: one custom-designed fixed-wing mothership, two child quadcopters, and a shared coordination layer with three live video feeds converging on one ground station.
// The platform
AeroCadastra’s current design pairs an original fixed-wing mothership with two child quadcopters. The key engineering challenge is integrating deployment, vehicle identity, telemetry and video interfaces so the fleet can operate as one system under a single operator.
// From imagery to maps
The AI baseline is YOLO26n fine-tuned on VisDrone, with an NVIDIA RTX 3050 used for training. The software baseline is Ubuntu 24.04 and ROS 2 Jazzy; onboard inference using Vicharak Axon is planned for a later stage. Mapping and detection performance must be measured before accuracy claims are made.
// Connected land records
The proposed solution further integrates these records into a unified digital land-information workflow:
The intended workflow can flag possible discrepancies for field verification; it does not independently establish legal ownership or cadastral boundaries.
The objective
To validate a scalable multi-UAV system for coordinated aerial coverage, unified telemetry and three-feed video, with potential applications in land survey, agriculture, disaster response, infrastructure inspection and environmental monitoring.
The initial proof of concept is one carrier, two child drones, one ground station and three video feeds. Larger fleets are a future goal, dependent on validation of the core interfaces.
- Concept DoneSystem architecture defined: one fixed-wing carrier and two autonomous child drones.
- Prototype airframe Baseline selectedCustom-designed carrier airframe and PETG for rigid/high-load printed parts; practical validation remains pending.
- AI perception baseline SelectedYOLO26n fine-tuned on VisDrone is the planned baseline; validated field performance is not yet claimed.
- Ground station In progressIntegrating fleet telemetry, three independent video feeds, AI perception and geospatial workflows in one operator station.
- Land records integration NextLinking RoR, mutation records and survey maps to precise geo-coordinates.
- Validation NextComplete bench tests and controlled flight tests; record measured results before making performance claims.
- Fleet scale-up NextMove from one carrier and two child drones to larger coordinated fleets.
System demonstration
Engineering baseline
AeroCadastra’s technical baseline includes a custom fixed-wing mothership, child quadcopters, ROS 2 software, AI perception and a unified video architecture.
The project is being developed by the AeroCadastra team at Amity School of Engineering and Technology for Smart India Hackathon 2026. No external partnership is asserted here.
Team members
Leads the project and ties the stack together: ground station, AI pipeline and hardware integration.
Led drone AI development and built AeroCadastra’s software and website, driving core technology and digital innovation.
AI pipeline: aerial person detection and tracking on drone imagery.
Drone electronics assembly, flight controller configuration, sensor integration, and hardware troubleshooting
Designed the Aerocadastra mothership, Child drones and Release mechanism, developed mechanical components, and assembled the drone structure and integrated components
Researched for AI , designed and assembly Aerocadastra mothership , desigened 3D components for Child drones and Release mechanism