Haptic-Based Complementary Filter for Rigid Body Rotations

Thesis— Prof. Ravi N. Banavar (IIT B), Prof. Domenico Campolo (NTU Singapore) 

The non-commutative nature of 3D rotations poses well-known challenges in generalizing planar problems to three-dimensional ones, even more so in contact-rich tasks where haptic information (i.e., forces/torques) is involved. In this sense, not all learning-based algorithms that are currently available generalize to 3D orientation estimation. Non-linear filters defined on SO(3) are widely used with inertial measurement sensors; however, none of them have been used with haptic measurements. This paper presents a unique complementary filtering framework that interprets the geometric shape of objects in the form of superquadrics, exploits the symmetry of SO(3), and uses force and vision sensors as measurements to provide an estimate of orientation. The framework’s robustness and almost global stability are substantiated by a set of experiments on a dual-arm robotic setup. 

Teaching Assistantship —  

1. Prof. Leena Vachhani's course on Embedded Control and Robotics

   • Assisted in designing assignments, conducting tutorial sessions, and assessing lab submissions for 40+ students.

2. Prof. Vivek Natarajan's laboratory course on Systems and Control Engineering

   • Designed an experiment for trajectory tracking of TurtleBot3 using Frenet Frame for 25+ students.

Planning Frameworks for Robotic Insertion Tasks 

Summer research under the guidance of Prof. Domenico Campolo, Director of RRC @NTU. This research is currently continued as my Master's Thesis at IIT Bombay under the guidance of Prof. Ravi N. Banavar.

• Studied difficulties in existing planning frameworks for robotic manipulators in contact-rich insertion tasks. 

• Identified and analyzed the problems with vision, learning, FEA, & potential fields approach in literature. 

• Explored Superquadric representation, Hydroelastic contact model, Black-Box Optimization, Dynamic Movement Primitives, and Drake simulator. 

• Developed an algorithm to model non-convex objects into multiple Superquadrics using Voxelized Hierarchical Approximate Convex Decomposition (V-HACD) library.

e-Yantra is a project sponsored by MoE, Government of India, through the National Mission on Education through ICT (NMEICT). Principal Investigator- Prof. Kavi Arya

Sr. Project Technical Assistant: 2020-2024

Research & Development:

◦ Designed an autonomous warehouse management and agricultural harvester system utilizing a mobile rover and UR5 arm, facilitating remote educational experiences for 1433 and 1312 students in real-world setting. 

◦ Developed an indoor localization method for Nano Aerial Vehicle (NAV) using off-board computing, monocular camera, and Whycon marker under 50 USD. 

Leadership & Management: 

◦ Trained 4,000+ students for e-Yantra Robotics Competition (eYRC) in 4 editions. 

◦ Spearheaded e-Yantra Innovation Challenge (eYIC) for 2 consecutive years, increasing the new colleges by 142%. 

◦ Conducted two-day workshop on Embedded Systems for over 100 faculties from engineering colleges. 

◦ Managed and mentored 15 interns for 5 Summer internship projects at IIT Bombay.

Intern: May 2018 - July 2018

◦ Developed Iterative Feedback Auto-tuning algorithm for Proportional Integral Derivative (PID) controller for drones. On average, it took 75 seconds for auto-tuning to complete. 

Automatic Umbrella Dryer 

Bachelor’s Project — Prof. Surendra Singh Rathod, Principal

Received research grant from the University of Mumbai for developing this project.

◦ Designed, simulated, and built 2 prototypes, including their mechanical & electronic components for automatic drying of foldable umbrellas in under 120 seconds. 

◦ Generated flow simulations to minimize heat loss, created custom gearbox using bevel gears, and implemented crank-shaft mechanism, including automatic clamping. 

◦ Developed the algorithm in Embedded C language using Texas Instrument’s MSP432 microcontroller.