Robotics · Computer Vision · Autonomy

Karteek Gandiboyina

MS Autonomy & Robotics, UIUC · GPA 3.89/4.0
gkarteek99@gmail.com

Building autonomy stacks for wildfire-suppression UAVs at Rational CyPhy Inc. Previously R&D Robotics Engineer at Konica Minolta, Tokyo. B.Tech EE from IIT Kharagpur.

1st Place · CVPR 2026 IROS 2026 2 Patents JAXA Collab
Karteek Gandiboyina

Research

01
3D Gaussian Splatting 6-DoF Pose Estimation SLAM Sim-to-Real Transfer World Models Vision-Language Models Aerial Autonomy
1st Place · CVPR 2026 FMEA Workshop
ENACT Embodied Cognition Challenge
CVPR 2026

Fine-tuned Qwen2.5-VL-7B with LoRA SFT on 7,672 samples. 94.6% Task Accuracy, 98.4% Pairwise Accuracy — outperforming GPT-5 and Gemini 2.5 Pro zero-shot. Invited contributed talk. [code]

Conference Paper
FalconTrack: Photorealistic Auto-Labeled Perception and Physics-Aware Vision-Based Aerial Tracking

Yan Miao, Karteek Gandiboyina, Noah Giles, Hideki Okamoto, Bardh Hoxha, Georgios Fainekos, Sayan Mitra

IROS 2026

Unified 3DGS-based simulator automating 6-DoF pose and mask labeling (10k images in <20 min), enabling zero-shot sim-to-real aerial tracking with 100% success on high-speed trajectories. [arXiv]

Patent — USPTO
Learning Data Generation Support Device, Movement Controller, and Article Acquisition System
US: 20250336175

Automated data generation framework eliminating manual labeling bottlenecks for complex 3D pick-and-place robotics.

Patent — JPO
Component Posture Information Acquiring Device and Posture Determination Method
JPO: 2023-060687

Vision-based sensing system for accurate component posture estimation during robotic pick-and-place.

Experience

02
Rational CyPhy Inc. Feb 2026 – Present
Robotics Engineer
  • Architected autonomy stack for NASA-aligned wildfire-suppression UAVs on VOXL 2, targeting BVLOS in GPS-denied environments.
  • Fused Basalt VIO with GPS; designed seamless sensor failover for degraded conditions.
  • Deployed real-time 3DGS for aerial environment reconstruction and situational awareness.
  • Prototyped self-organizing multi-drone architectures for comms/GPS outage resilience.
Coordinated Science Lab, UIUC — Prof. Sayan Mitra Sept 2025 – Jan 2026
Graduate Research Assistant
  • Trained DreamerV3 World Model on Crazyflie FPV data for imagination-based policy training.
  • 6-DoF pose estimation pipeline using 3DGS-generated synthetic datasets (10k frames) + fine-tuned U-Net; sub-centimeter accuracy in sim-to-real.
  • Actor-critic (PPO) policy within latent world model; waypoint completion 70% → 95%.
  • Min-snap trajectory planners with differential flatness, reducing planning time by 30%.
Konica Minolta — R&D HQ, Tokyo Jul 2021 – Aug 2024
R&D Robotics Engineer
  • Structured-light point cloud sensor: 0.99 IoU on 1 cm² items at 0.05 m under <10 Lux for bin-picking.
  • Auto-annotation pipeline (SAM, Detectron2, YOLOv7) reducing manual labeling >80%.
  • Language-conditioned multi-task RL (SAC + DAgger) with Georgia Tech; zero-shot task success +200%.
  • Collaborated with JAXA on 6-DoF grasping for multi-limbed robots aboard the ISS.
  • Filed 2 patents in computer vision and robotic grasping (USPTO & JPO).
Philips Innovation Campus — Bangalore Apr 2020 – Jul 2020
Machine Learning Intern
  • 3D residual U-Net for pulmonary nodule detection on 888 CT scans (LUNA16); FROC 0.914, +8% over baseline.

Education

03
University of Illinois Urbana-Champaign Aug 2024 – Dec 2025
MS in Autonomy & Robotics — GPA: 3.89/4.0
Deep Learning for Graphs · Computer Vision · Safe Autonomy · Deep Generative Models
IIT Kharagpur Aug 2017 – May 2021
B.Tech. in Electrical Engineering — GPA: 3.48/4.0
Deep Learning · Machine Learning · Embedded Systems · Control Systems

Projects

04
Drone Racing
Autonomous Drone Racing

MPC & PID spline tracker for gate navigation. NanoSAM keypoint detector for misaligned gate correction.

Record: 50.13s (Tier-1) · Error: 0.05m
MPCPIDNeRF AirSimNanoSAM
Code
VLM4Autonomy
VLM4Autonomy

SAM2 + optical flow + VLMs for real-time tracking and ego-vehicle motion estimation.

SAM2YOLOv8OpticalFlow SfMVLMs
Code
Stock Predictor
Volatility-Aware Stock Prediction

Transformer-VAE with RealNVP normalizing flows conditioned on VIX.

Wasserstein: 0.0629 · CosSim: 0.9072
PyTorchVAE RealNVPTransformers
Code

Skills

05
Vision
3D Gaussian Splatting NeRF SLAM 6-DoF Pose YOLO SAM Point Clouds Optical Flow SfM
Robotics
ROS/ROS2 MPC PID Min-Snap VIO Quad Control Manipulation AirSim MuJoCo
ML / DL
PyTorch JAX Transformers VAEs Diffusion GNNs DreamerV3 PPO / SAC VLMs
Tools
Python C++ Docker Linux Git VOXL 2