Research

From a broad perspective, my research areas of interest involve working with the various facets and principles of Engineering, Computer Science, and Mathematics. More specifically, my focus includes the areas of Autonomous Systems/Robotics, Mechatronics & Sensor Technology, Oceanography & Environmental Science, Simulation & Data Analysis, Computer Vision, and Applied Machine Learning.

  • Robotics, Mechatronics, and Sensors-based Technology: For this research area, my work involves both the hardware and software domain of engineering. A typical project in this category could involve various tasks starting from the creation of a hardware architecture (viz. a 3D-printed prototype) to deploying such a finished system with an array of onboard sensors controlled by an embedded architecture (viz. a microcontroller or a microprocessor) in a real-world environment. Such systems can complete tasks either with the aid of commands from telemetry devices or by following an autonomous logic.
  • Oceanography, Autonomous Navigation, and Environmental Monitoring: This is one of the major application-focused areas that I am interested in. Although these terminologies might seem unrelated at first glance, they are inter-dependent for most application-centered tasks as we require autonomously navigating systems that can independently carry out monitoring operation in remote and/or hostile environments of various ecosystems, such as deep beneath the oceanic water surface. By deploying unmanned systems with sensory integrations in remote locations, it is possible to procure research data from such environments which can then be analyzed for further study.
  • Simulation and Environmental Data Analysis: Testing out prototype models or even analytical theories in simulation platforms is always useful to learn more about unknown systems. Various software programs are used for this purpose in my studies (viz. Blender, Rviz, Gazebo, etc.). Data analysis tasks for environmental applications involve either publicly available databases (such as from USGS and EPA repositories), or sample data collected using unmanned vehicles. Such data could be analyzed using various kinds of learning algorithms that can help us to learn more about the region, or the level of environmental degradation of an ecosystem.
  • Computer Vision: With the help of suitable machine learning and pattern recognition algorithms, I am interested in learning how computer vision techniques could better be applied for differentiating between various types of image datasets, either for remote sensing based environmental applications (viz. recognizing various terrain-types in satellite images), or medical procedures (viz. discriminating between benign and malignant tumor formations).

Please refer to my LinkedIn page for more details.