Ajinkya Gorad
Machine Learning Researcher | Sensor Fusion | AI-Driven Systems
Doctoral Candidate, Aalto University
Summary
Applied researcher with a Ph.D. in Electrical Engineering and Automation, specializing in image-based sensing, sensor fusion, and AI-driven system design. Interests span embedded systems, health monitoring, AR dashboards. Led and contributed to ESA NAVISP - funded projects with real-time multi-modal perception for maritime awareness - classical signal processing and ML image processing techniques, enabling deployment-ready solutions from concept to prototype. Participated in numerous projects spanning algorithm, software, embedded development with focus on system design and hardware integration. My recent focus have been on AI robotics which involves both awareness and action.
Experience
Doctoral Researcher
Aalto University | 2019 – Present | Espoo, Finland
- Developed classical and deep-learning-based sensor processing systems, for healthcare, speech, & maritime.
- Led European Space Agency NAVISP funded projects in navigation using AI.
- Designed dashboards using LightningChart for Python/JS and AR interfaces.
- Trained and tested Vision language action AI models on Lerobot robotic arm.
- Conducted full-stack ML projects: data collection, preprocessing, training, benchmarking.
Research Interests
- Health monitoring devices
- Sound and imaging sensors for environmental awareness
- Spiking neural networks and liquid computation
- Space exploration
- Unifying knowledge structures
- Navigation & Earth Observation
- Image Processing
- Maritime & Healthcare Applications
- Machine Learning & Artificial Intelligence
- Consciousness & Awareness Studies
Publications
Publications
Ice-track detection using deep networks in visible and infrared from ship-borne camera
Under review. Preprint available at SSRN, 2024
View on SSRN
Vessel Bearing Estimation Using Visible and Thermal Imaging
SCIA 2023, Springer
Rao-Blackwellized Monte Carlo Data Association With Deep Metric For Object Tracking
MLSP 2023, IEEE
Bearing estimation using foghorn sounds
Applied Acoustics, 2025, Elsevier
Machine learning and state-space methods for healthcare, speech, and maritime awareness
A Gorad
Aalto University, Doctoral Thesis, 2025.
Hardware-Friendly Synaptic Orders and Timescales in Liquid State Machines for Speech Classification
IJCNN 2021, IEEE
View on IEEE Xplore
Parameter Estimation in Non-Linear State-Space Models by Automatic Differentiation of Non-Linear Kalman Filters
MLSP 2020, IEEE
View on IEEE Xplore
Respiratory Pattern Recognition from Low-Resolution Thermal Imaging
ESANN 2020
View PDF
Predicting Performance using Approximate State Space Model for Liquid State Machines
IJCNN 2019, IEEE
View on IEEE Xplore
Analytical estimation of LER-like variability in GAA Nano-sheet transistors
VLSI-TSA 2019, IEEE
View on IEEE Xplore
Oscillatory 2-neuron sub-network design and performance based on sub-threshold CMOS operation
ICEE 2018, IEEE
View on IEEE Xplore
Key Projects
Maritime Awareness Systems
European Space Agency NAVISP / Finnish Geospatial Institute / Fleetrange Oy / Aalto University
Multi-modal ship detection and navigation using visual-thermal camera fusion, foghorn-based acoustic bearing estimation, and semantic segmentation of sea ice tracks. Developed under the ESA Navigation Innovation and Support Programme (NAVISP) to enhance situational awareness in polar maritime environments.
Lerobot SO100-SO267
Self project
Trained Vision-Language-Action models for real-time 3D printed robotic arm control.
Speech Classification using Spiking Neural Networks
IIT Bombay
Liquid State Machine architecture with cochlear-inspired input and memory metric τM for isolated word classification.
DeepRBMCDA object tracker and Differentiable Kalman Estimation
Aalto University
Probabilistic multi-object tracking using Deep Rao-Blackwellized Monte Carlo particle filter with ReID integration; differentiable Extended Kalman Filter for parameter learning in nonlinear systems.
Physiological Monitoring
Sense4Health / Aalto University
Non-contact respiratory monitoring via nasal thermal imaging, IMU-based breathing estimation, and inertial sensor-based heart rate detection in embedded systems.
Skills
Languages
English
Marathi (Native)
Finnish (Learning)
Programming
Python
MATLAB
C/C++
Embedded C
Java
JavaScript
Bash
SQL
Hardware Platforms
Tools & Environments
Technical Skills
Prototyping Highlights
AR Integration: Augmented reality dashboards for data visualization
Supported by Funding
Aalto Doctoral Funding (2020–2023)
ESA NAVISP Funding (Maritime-AI Nav, ENHANCE)
AI Research Funding (AI-MODE, MR-HISTO)
Supervision Experience during Ph.D.
2020–2022: Supervised BSc students on non-contact heart monitoring and IMU-based motion analysis
Societal impact
Dataset: Baltic Sea Ice Dataset: GNSS+thermal imagery with annotations for autonomous marine systems.
Media: YLE News 2023: AI-enhanced maritime navigation.
Doctoral Thesis: "Machine learning and state-space methods for healthcare, speech and maritime awareness"
Doctoral dissertation open access | Permanent link: URN:ISBN:978-952-64-2543-6
Hobbies & Other Interests
Philosophical Thinking: Reflects on the nature of life, consciousness, and systems; writes abstract explorations.
Abstract Writing & Blogging: Maintains personal notes and conceptual essays on thought, design, and cognition.
Plant Cultivation: Enjoys nurturing indoor plants and observing biological growth patterns.
Medical Context: Has a spinal implant following a major road accident in 2024; currently fully active with precautions.