Kartikeya Singh Ph.D. Candidate in Computer Science

About Me

Kartikeya Singh

I am a Ph.D. candidate in Computer Science at the University at Buffalo, working in the Distributed Robotics and Networked Embedded Sensing (DRONES) Lab under Prof. Karthik Dantu. I am currently a Research Scientist Intern at Mitsubishi Electric Research Laboratories (MERL), Cambridge, where I work with Dr. Stefano Di Cairano .

My research focuses on enabling legged robots to navigate complex and unseen terrains using uncertainty-aware perception and planning. I develop learning-based methods for predicting future footholds, modeling epistemic uncertainty, and generating feasibility-aware costmaps for robust navigation.

More broadly, I am interested in bridging perception, control, and planning to build autonomous robotic systems that can reliably operate in real-world environments. My work combines simulation and real-world experiments on platforms such as quadruped robots, with an emphasis on generalization and safety in out-of-distribution scenarios.

Latest News

Jan 2026

Joined MERL as Research Scientist Intern

Started working at Mitsubishi Electric Research Laboratories (MERL), Cambridge, focusing on uncertainty-aware perception and planning for legged robots.

May 2024

Our Paper Accepted at Workshop in Field Robotics at ICRA 2025

Recognized as one of the top 10 emerging researchers at the Workshop on Field Robotics during ICRA 2025 in Atlanta, Georgia

May 2024

Represented SUNY Buffalo at AI expo in Albany, NY

Honored to connect with leaders like SUNY Chancellor Dr. John B. King (former U.S. Secretary of Education) and state policymakers.

April 2024

Paper got accepted at ICRA 2024

L-DYNO framework nominated for best paper in robotic perception category.

Publications

CART Demo

CART: Context-Aware Terrain Adaptation for Stable Locomotion

K Singh, Y Kim, Y Turkar, K Dantu
RaL 2025 (Under Review) June 2025
Culvert Inspection

Autonomous Culvert Inspection on the Erie Canal using Legged Robots

K Singh, Y Turkar, Y Kim, M Lengel, K Dantu
ICRA 2025 (WFR) May 2025
PANOS

PANOS: Payload-Aware Navigation in Offroad Scenarios

K Singh, Y Turkar, C Alukal, K Dantu
Nerc 2024 May 2025
L-DYNO Demo

L-DYNO: Framework to Learn Consistent Visual Features Using Robot's Motion

K Singh, C Adhivarahan, K Dantu
ICRA 2025 April 2025
OFFSEG Demo

OFFSEG: A Semantic Segmentation Framework for Off-Road Driving

K Viswanath*, K Singh*, P Jiang, PB Sujit, S Saripalli
CASE 2021 August 2021

Research and Teaching Experience

Research Scientist Intern

Mitsubishi Electric Research Laboratories (MERL), Cambridge, MA Jan 2026 – Present
  • Researching uncertainty-aware costmaps and task-conditioned perception for legged robot navigation
  • Designing learning-based models for robust foothold prediction and out-of-distribution terrain awareness
  • Bridging perception and planning for reliable deployment of autonomous robots in challenging environments

Research Assistant

University at Buffalo 2023–Present
  • Developed PANOS for legged robot navigation in unstructured environments
  • Created L-DYNO framework for visual feature learning using robot motion
  • Implemented terrain adaptation algorithms for quadruped robots

Teaching Assistant

University at Buffalo 2022–2023
  • Robotic Algorithms (CSE 568/468)
  • Data Intensive Computing (CSE 587)
  • Developed course materials and supervised student projects

Research Associate

IISER Bhopal 2020–2022
  • Developed OFFSEG framework for off-road semantic segmentation
  • Implemented multi-modal sensor fusion techniques
  • Contributed to autonomous navigation systems for agricultural robots

Contact

Google Scholar

Scholar Page

Office

Davis Hall 106, North Campus, University at Buffalo

LinkedIn

Link