🎓 Mechatronics Engineering undergraduate | 🤖 Researcher in Intelligent Systems Under Resource Constrain 📍 Nigeria

I am an undergraduate researcher focused on intelligent autonomous systems under resource constraints. My work bridges practical engineering and applied AI, aiming to design systems that adapt efficiently and robustly in complex, dynamic environments.

• Design, control, and learning of autonomous systems
• Resource- and compute-efficient AI for robotics and embedded systems (TinyML, parameter-efficient fine-tuning, model compression)
• Sim-to-real system testing: from simulation to real hardware deployment
• Research skill development, academic writing, and reproducible experimentation

My research focuses on behavioral adaptation and efficient computation under resource constraints, aiming to maintain robustness, efficiency, and task specialization in low-power, low-memory, or latency-critical environments. Key areas include:

1. Adaptive & Intelligent Control

   • Systems that adjust behavior autonomously in real-time using adaptive or learning-based control.

2. Embedded Machine Learning / TinyML

   • Deploying models in low-resource environments (edge devices, IoT, robots).
   • Techniques: quantization, pruning, knowledge distillation, lightweight architectures (MobileNet, SqueezeNet).

3. Parameter-Efficient Fine-Tuning

   • Adapting large models to new tasks with minimal computation using LoRA, adapters, and prompt/prefix tuning.

4. Efficiency under Constraints

   • Optimizing compute, memory, energy, and latency for real-world autonomous systems.

• Compute / Memory: How small and efficient can models be? Can they run on embedded systems?
• Energy / Power: Can the system operate under limited battery or power budgets?
• Latency / Real-Time Performance: Does the system respond fast enough for its tasks?
• Robustness: Can it handle sensor noise, dynamic changes, or unexpected conditions?
• Behavioral Adaptation: Can the system autonomously adjust policies or control parameters under changing environments?

These are the tools and frameworks I actively use to design, test, and deploy efficient, adaptive autonomous systems:

• Programming & Computation: Python, C/C++ – for implementing adaptive control algorithms, AI models, and embedded systems.
• Embedded & Edge Deployment: TinyML frameworks, microcontroller programming – for deploying AI models under low-power and memory-constrained conditions.
• Machine Learning & Model Efficiency: PyTorch, TensorFlow – for building, fine-tuning, and compressing deep learning models (LoRA, adapters, pruning, quantization, knowledge distillation).
• Simulation & Testing: MATLAB/Simulink, Gazebo – for simulating control systems, autonomous behaviors, and system adaptation before real-world deployment.
• Experimentation & Reproducibility: Git, Linux, Docker – for version control, development environments, and containerized experiments ensuring reproducibility.

This GitHub documents my learning, experimentation, and research-oriented projects.
Repositories include simulations, adaptive control experiments, embedded AI projects, and applied programming for autonomous systems.

My research interests and skills are actively evolving as I explore robotics, embedded AI, and adaptive autonomous systems. This space will grow with my work and experimentation.

• LinkedIn: https://linkedin.com/in/glory-bagai
• Email: bagaiglory@gmail.com