01
VTOL flight platform
A vertical take-off, fixed-wing testbed: airframe, embedded flight logic, and the instrumentation to learn from every flight.
Autonomous systems R&D — Haridwar, India
Machines for the physical world
Brobot is an early-stage deep-tech company building autonomous aerial platforms, swarm coordination software, and embedded intelligence. We are in our research phase — and we are building it in the open.
- Founded
- 2025
- Stage
- R&D phase
- Base
- 29.9457° N / 78.1642° E
- Focus
- Autonomous systems
- Platform
- VTOL UAV & Robots
- Software
- ROS 2 swarm stack
Current focus
Current focus
Three engineering threads, one system. Everything we build today is groundwork for coordinated autonomous fleets.
02
Swarm coordination
A ROS 2 multi-agent stack for decentralised task allocation and inter-vehicle communication — built and tested in containerised simulation first.
03
Geospatial intelligence
GeoAI pipelines that turn what our machines see into maps and decisions: terrain awareness, 3D reconstruction, and mission planning.
Research areas
Where we build depth
4 active threads, honestly labelled. Nothing graduates until it survives the field.
- active
Autonomous navigation
Getting a vehicle from A to B without a human in the loop: path planning, obstacle handling, and behaviour under degraded GPS.
- active
Swarm & multi-agent systems
Decentralised coordination for fleets: inter-agent communication, task allocation, and emergent collective behaviour across distributed vehicles.
- active
UAV & VTOL platforms
The aircraft themselves: airframe design, flight-control logic, propulsion sizing, and the embedded systems that keep them in the air.
- active
Geospatial intelligence (GeoAI)
Turning sensor data into spatial understanding: mapping, terrain analysis, and location-aware decision-making for mission planning.
- exploratory
Computer vision & perception
What the machine sees: detection, tracking, and the perception stack that feeds navigation and mapping.
- exploratory
SLAM & 3D mapping
Simultaneous localisation and mapping: building 3D representations of unknown environments while navigating them.
Capability map
Engineering expertise
- Robotics middleware
- ROS 2 nodes, DDS pub/sub, services, containerised sim
- Multi-agent systems
- Decentralised coordination, task allocation, swarm behaviour
- Embedded & UAV
- Flight control logic, firmware, hardware prototyping
- Applied ML
- Perception, GeoAI, data pipelines, model integration
- Software products
- Cross-platform apps shipped to production (Flutter)
- Systems
- Docker, Linux, networking, CI-driven development
Tooling
Technology stack
Autonomy
- ROS 2
- Gazebo/SITL simulation
- Docker
- Python
- C++
Intelligence
- PyTorch-based ML
- Computer vision
- GeoAI / geospatial
- Edge inference
Hardware
- Embedded C/C++
- Flight controllers
- Sensor integration
- Rapid prototyping
Product
- Flutter / Dart
- TypeScript
- Next.js
- Cloud APIs
Development roadmap
The plan we operate against
Sequential phases — each produces the evidence the next one needs. Dates are targets and say so.
01● now
Research
2025 — 2026
02
Prototype
2026
03
MVP
2026 — 2027
04
Pilot deployment
2027
05
Commercial launch
2027
06
Scaling
2028+
07
Global expansion
2029+
Featured projects
On the bench right now
Flight platform
prototypeVTOL flight testbed
A vertical take-off, fixed-wing UAV built in-house — the airframe on which every layer of our autonomy stack gets proven.
2025 — ongoing
Autonomy software
researchMulti-agent coordination stack
A ROS 2-based software stack for decentralised multi-vehicle coordination — communication, task allocation, and collective behaviour, validated in containerised simulation.
2026 — ongoing
Intelligence
researchGeoAI terrain intelligence
Geospatial intelligence workflows that turn raw spatial data into maps, terrain awareness, and mission-planning inputs for autonomous flight.
2026 — ongoing
Latest updates
From the log
Company
brobot.dev goes live
Research
R&D internship at IISER Bhopal begins
Milestone
National hackathon: team led end-to-end
Engineering blog
Recent writing
2 min read
Why we are building in the open, starting from zero
Brobot is a research-stage company with no product to sell yet. Here is why our site says so — and why we think radical honesty is the right strategy for a deep-tech startup.
2 min read
Notes on multi-agent coordination with ROS 2
Field notes from hands-on work with ROS 2 multi-agent systems: what the middleware gives you for free, where coordination actually gets hard, and why containerised simulation is non-negotiable.
Founder
Yuvraj Gargvanshi
Third-year B.Tech (AI & ML) student and robotics developer, building Brobot from first principles.
Yuvraj is a third-year B.Tech Computer Science (AI & ML) student at Haridwar University. He interned at IISER Bhopal as an R&D intern in UAV robotics, swarm intelligence, and autonomous systems — working hands-on with ROS 2, Docker, and multi-agent robotic communication frameworks, alongside applied machine learning and embedded UAV development.
Full profileStraight answer
Why take us seriously
We are early, and we say so. What we offer instead of traction is evidence of execution: live shipped software, an R&D internship in multi-agent robotics at a national research institute, and a public engineering log you can audit.
Every claim on this site maps to something concrete — a prototype, a repository, a test, or a written note. If it is a plan rather than a result, it is labelled as one.