Designing Autonomous Systems and Intelligent Machines for Real-World Impact
Autonomous Systems · Robotics · Aerospace · AI-Driven Engineering
Engineering student focused on robotics, autonomous systems, and aerospace technologies. I design and build complex systems integrating AI, control theory, and hardware to solve real-world problems.
I'm Yug Umasudhan — an engineering student operating at the intersection of autonomous systems, robotics, and aerospace. My work isn't about individual components; it's about integrated systems that perform in the real world.
I approach every project as a systems engineering challenge — understanding how mechanical design, embedded electronics, control algorithms, and AI perception layers must cohere into a single optimized solution. My focus is on building things that work outside the lab.
I approach engineering as a systems problem — integrating mechanical, electrical, and software components into cohesive, optimized solutions. My work emphasizes real-world performance, autonomy, and scalability.
A Cognitive Autonomous Underwater Vehicle designed for real-time multi-biomarker sensing, AI-driven environmental analytics, and precision biotic remediation. Merging marine robotics with environmental intelligence.
An intelligent autonomous system integrating edge computing, adaptive navigation, and real-time environmental analysis for high-precision surveillance and response. Designed to operate in denied or degraded environments.
Full-scale electric autonomous vehicle with Jetson Orin compute, LiDAR + camera sensor fusion, neural network steering via imitation learning, and ROS 2 architecture.
EEG-based brain-computer interface with real-time intent decoding via ML classifier. Servo-actuated mechanical arm driven by neural signal processing.
Custom-built quadcopter with IMU-based stabilization and advanced flight control systems. Focus on embedded flight controller design.
Target speed: Mach 1.5+. Custom avionics with PCB design, real-time telemetry transmission via LoRa radio. Full avionics stack.
CO₂/nitrogen propulsion system with PID-based attitude control and full microcontroller integration. Research-grade propulsion testbed.
Thrust Vector Control (TVC) with full GNC system and autonomous landing sequence. End-to-end aerospace systems project spanning guidance, navigation, control, and recovery.
Systems-level architecture. Define requirements, constraints, and interfaces before touching hardware or code. Every subsystem is designed with integration in mind.
Build minimum viable systems that expose the hardest problems first. Physical hardware reveals what simulation cannot. Fail fast, fail cheap.
Real-world validation over theoretical performance. Data from actual operation is the only ground truth. Instrumented testing with logged telemetry on every system.
Engineering is convergence through iteration. Each cycle produces quantifiable improvement. Version control isn't just for software — it's for hardware too.
I build systems, not components. The discipline of systems engineering forces you to think about interfaces, failure modes, and emergent behavior before they become problems in the field. My focus is always on the whole — how mechanical structures, embedded electronics, control algorithms, and AI perception layers combine into a coherent, reliable machine.
Every component decision affects the whole. Design for integration, not isolation.
Simulation is a starting point. Physical testing is where engineering earns its credibility.
Building systems that operate without human intervention — that's the hard, interesting problem.
Open to discussing engineering collaborations, project feedback, research opportunities, and technical conversations. If you're working on interesting problems in robotics, aerospace, or autonomous systems — reach out.