Omey Mohan Manyar

Omey Mohan Manyar

Ph.D. Student in Robotics

University of Southern California

I am a Ph.D. student in robotics at USC, affiliated with the Realization of Robotics Systems Lab (RRoS) advised by Dr. Satyandra K. Gupta. My research is centered around integrating physics-informed learning methods into robotics, focusing on enhancing robots’ capability to manipulate objects with complex physics. I’m developing practical and scalable approaches to create explainable models that exploit physics-aware inductive biases for robots to perform complex tasks.

Before starting my Ph.D., I have led multiple industry-funded projects at USC as a Master’s student. In the past, I worked for Rolls-Royce, Singapore, as a robotics engineer for two years.

Download my CV .

Interests
  • Robot Learning
  • Robotic Manipulation
  • Deep Learning
  • Motion Planning
Education
  • PhD in Mechanical Eng. (Robotics), 2025

    University of Southern California

  • MSc. in Mechanical Eng., 2021

    University of Southern California

  • Bachelor of Technology, 2016

    National Institute of Technology Karnataka

News

Experience

 
 
 
 
 
Amazon Robotics
Applied Science Intern
May 2023 – Aug 2023 Seattle, WA
Worked with the perception team in the Stow Manipulation Project
 
 
 
 
 
Realization of Robotics Systems Lab
Research Assistant
Realization of Robotics Systems Lab
Oct 2019 – Present Los Angeles, CA
Team lead for industry funded projects for developing advanced robotic solutions
 
 
 
 
 
Rolls-Royce Plc
Robotics Technologist
Rolls-Royce Plc
Jan 2018 – Jul 2019 Singapore
Worked with research partners to mature technology beyond TRL 4
 
 
 
 
 
General Motors India Pvt. Ltd.
Graduate Engineer Trainee
General Motors India Pvt. Ltd.
Aug 2016 – Jul 2017 India
Selected as a part of the flagship campus connect program by GM India

Publications

Search for a specific publication filtering publications.
(2023). Physics-Informed Learning to Enable Robotic Screw-Driving Under Hole Pose Uncertainties. IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

PDF Cite

(2023). Inverse Reinforcement Learning Framework for Transferring Task Sequencing Policies from Humans to Robots in Manufacturing Applications. IEEE International Conference in Robotics and Automation.

PDF Cite Code

(2022). Physics Informed Synthetic Image Generation for Deep Learning based detection of wrinkles and folds. ASME Journal of Computing and Information Science in Engineering.

PDF Cite Code Dataset

(2022). Synthetic image-assisted deep learning framework for detecting defects during composite sheet layup. ASME IDETC-CIE Conference, 2022.

PDF Cite Code Dataset

(2022). Visual servo-based trajectory planning for fast and accurate sheet pick and place operations. ASME Manufacturing Science and Engineering Conference.

PDF Cite DOI

(2021). A Simulation-Based Grasp Planner for Enabling Robotic Grasping during Composite Sheet Layup. IEEE International Conference in Robotics and Automation.

PDF Cite DOI

(2021). A Digital Twin for Automated Layup of Prepreg Composite Sheets. ASME Journal of Manufacturing Science and Engineering.

PDF Cite DOI

(2021). A Digital Twin for Automated Layup of Prepreg Composite Sheets. ASME Manufacturing Science and Engineering Conference.

PDF Cite

(2021). Toolpath Generation for Robot Filleting. Internatoinal Conference on Advanced Surface Enhancement.

PDF Cite

(2021). An adaptive framework for robotic polishing based on impedance control. The International Journal of Advanced Manufacturing Technology.

PDF Cite DOI

(2019). In-process virtual verification of weld seam removal in robotic abrasive belt grinding process using deep learning. Robotics and Computer-Integrated Manufacturing.

PDF Cite DOI

(2018). An AWS Machine Learning-Based Indirect Monitoring Method for Deburring in Aerospace Industries Towards Industry 4.0. Applied Sciences.

PDF Cite DOI

Projects

*
Inverse Reinforcement Learning for Task Sequencing
Can we have robots learn an expert’s policy to perform task sequencing for complex processes such as surface finishing or composite layup? This project attempts to answer this question by using the inverse reinforcement learning framework.
Inverse Reinforcement Learning for Task Sequencing
Multi-Robot Cell for Automated Composite Layup on a Rotating Tool
Composite manufacturing is an industry that is growing rapidly. In this project, we collaborated with Advanced Technology Group at Lockheed Martin to design a multi-robot cell for sheet manipulation, draping and heating. No programming is required and an easy to use interface enables quick setup times with 30% reduction in touch time.
Multi-Robot Cell for Automated Composite Layup on a Rotating Tool
A Multi-Arm Robotic Assistant for Sheet Manipulation and Draping
Sheet manipulation for highly complex tasks such as prepreg composite layup is complex and challenging. The sheets used in such processes are large in size and difficult to manipulate using a single robot. In this project, the focus was to develop a multi-robot assistant system for aiding an operator in composite sheet layup task.
A Multi-Arm Robotic Assistant for Sheet Manipulation and Draping
ADAMMS-UV Robot built for fighting COVID
Disinfection is an important process for fighting a global pandemic. This work leverages a mobile base to perform uv-based disinfection of indoor spaces and hard to reach places such as cabinets and drawers.
ADAMMS-UV Robot built for fighting COVID

Accomplish­ments

Best Paper Award at IDETC-CIE 2022
Viterbi Graduate Student Fellowship
Academic Excellence Award
NSF Travel Award for ASME MSEC & IDETC-CIE Conferences (2021, 2022)