Harshitha Belagavi Rajaprakash ☕️
Harshitha Belagavi Rajaprakash

Robotist, Researcher

About Me

Hi There! :)

I am a first year PhD student, advised by Professor Jesse Thomasan at University of Southern California, LA. I am interested in building robots that interact with humans and effectively collaborate and assist humans.

Previously, during my MSCS at USC, I was a research assistant with Prof Shri Narayanan at the SAIL Lab and at SLURM Lab with Prof Daniel Seita on robotic manipulation. During my time at Indian Institute of Science(IISc), Bangalore, India; I worked as a research assistant with Prof Pradipta Biswas at I3D Labs.


Download CV
Interests
  • Robotics
  • Computer Vision
  • Robotic Perception
  • Assistive Robotics
  • Machine Learning
  • Artificial Intelligence
Education
  • PhD Computer Science

    University of Southern California

  • MS Computer Science

    University of Southern California

  • BE Computer Science and Enginering

    University of Visvesvaraya College of Engineering

📚 My Research

SLURM Lab, USC

  • Working at Sensing, Learning, and Understanding for Robotic Manipulation (SLURM) Lab at University of Southern California (USC) on Caregiving Robots, Plant Manipulation.
  • Working on robotic manipulation in contact-rich scenarios and leveraging trajectory optimization for planning.
  • Building a simulation for scooping particular particles in a clustered environment.
  • Building a solution to enable robotic assistance for dressing patients with muscular dystrophy and other neuro-immune conditions, as well as working on assistance for sponge bathing a patient to ensure maximum coverage.

I3D Lab, Indian Institute of Science, Bangalore

  • Worked at Intelligent Inclusive Interaction Design (I3D) Lab, CPDM, Indian Institute of Science, Bangalore.
  • Performed a comparative study across Mixed Reality and Virtual Reality-based remote welding Digital Twin applications and mapped the coordinates defined for welding movement to the robot coordinates using a regression technique.
  • Built an assistive Assembly Process Instruction tool in collaboration with Collins Aerospace, where Computer Vision is used to detect the components.
  • Built a custom dataset for the components in hand and gained experience in the various pre-processing steps involved with creating the dataset, using both real and synthetic images. Tried different augmentations on the dataset created.
Featured Publications
Recent Publications
(2024). Investigating Swimming Effect of Holograms in Mixed Reality.
(2024). Investigating Swimming Effect of Holograms in Mixed Reality. In International Conference on Pattern Recognition(ICPR) 2024.
(2024). I-ROD: An Ensemble CNNs for Object Detection in Unconstrained Road Scenarios. Signal, Image and Video Processing Journal, Springer; (Accepted, November 2024).
(2024). Investigating Swimming Effect of Holograms in Mixed Reality. International Conference on Pattern Recognition(ICPR), 2024; (Accepted).
(2024). Harnessing Learn Rate Schedule for Adaptive Deep Learning in LoRaWAN-IoT Localization. IEEE Access.