Presentations & Posters |

SPIE Defense + Security Poster

Tue, 28 Apr 2026 00:00:00 +0000

Poster presentation describing a Physics-Informed Attentive Neural Process (PI-AttNP) framework for multi-time-scale vehicle state estimation using Automatic Identification System (AIS) ship-tracking and Automatic Dependent Surveillance–Broadcast (ADS-B) aircraft-tracking datasets. The approach incorporates simplified physics-based motion models directly into an attentive neural process architecture to improve predictive accuracy across widely varying temporal resolutions.

Experimental evaluation demonstrated improved estimation performance relative to several state-of-the-art neural and probabilistic estimation baselines while maintaining real-time inference capability. The work was accepted for both paper + presentation at SPIE Defense + Security 2026.

Real-Time Performance Analysis of Multi-Fidelity Residual Physics-Informed Neural Process-Based State Estimation for Robotic Systems

Fri, 27 Mar 2026 00:00:00 +0000

This presentation introduces the Multi-Fidelity Residual Physics-Informed Neural Process (MFR-PINP), a novel probabilistic machine learning framework for real-time state estimation of uncertain robotic systems. The proposed approach combines low-fidelity inertial and encoder measurements with higher-fidelity LiDAR and RGB-D sensing through a residual neural process architecture that incorporates physical motion models directly into the learning process.

Experimental validation was performed on a physical skid-steering robotic platform, demonstrating improved estimation accuracy, accelerated convergence, and real-time deployment feasibility relative to multiple baseline state estimation approaches. The framework additionally incorporates Split Conformal Prediction to provide uncertainty-aware state estimates with statistical coverage guarantees.

Neural Process-based Reactive Controller for Autonomous Racing

Thu, 05 Mar 2026 00:00:00 +0000

This presentation introduces a neural process-based reactive control framework for autonomous racing using an Attentive Neural Process (AttNP) and a physics-informed extension, the PI-AttNP. The proposed approach learns a Follow-The-Gap (FTG) navigation policy directly from LiDAR and vehicle dynamics data while incorporating an approximate FTG-derived control prior to improve convergence, prediction accuracy, and generalization in high-speed racing environments.

Experimental validation was performed in a simulated F1TENTH Ackermann steering racing environment, demonstrating improved control prediction performance and reduced collision frequency relative to AttNP and residual MLP baselines. The framework additionally incorporates a Control Barrier Function (CBF)-based safety filter that provides real-time collision avoidance guarantees while maintaining competitive lap completion performance.

Autonomous Navigation with ROS Navigation Stack

Fri, 21 Apr 2023 00:00:00 +0000

This technical presentation provides a complete walkthrough of a custom autonomous navigation stack developed for a differential-drive mobile robot using ROS Noetic. The presentation covers robot kinematics, sensor integration, and software architecture for a robotic platform equipped with a 360° LiDAR, Intel RealSense D435 RGB-D camera, BNO055 9-DOF IMU, and wheel encoders.

Topics include Extended Kalman Filter-based odometry fusion, RTAB-Map visual-inertial SLAM, AMCL particle-filter localization, map generation, and deployment of the ROS1 Navigation Stack for fully autonomous navigation within a known environment. The presentation emphasizes practical implementation details and system integration challenges encountered during development.

MIT Robust Robotics Group Internship Presentation

Sun, 05 Jul 2020 00:00:00 +0000

Research presentation summarizing work completed through the MIT Summer Research Program (MSRP) program within the Robust Robotics Group. The project investigated whether autonomous navigation could be achieved using only a single low-cost HC-SR04 ultrasonic range sensor combined with feedback control.

A custom wheeled robotic platform was developed using Arduino-based control hardware and evaluated against human teleoperation on a structured obstacle course. Experimental results demonstrated navigation performance within approximately 7% of human-operated performance metrics despite the robot relying on a drastically reduced sensing suite. The project highlights the potential for low-cost autonomous systems operating with minimal sensing requirements.

UNF SOARS Best Poster Award

Wed, 08 Apr 2020 00:00:00 +0000

Award-winning poster presentation describing research conducted within Stanford University’s Collaborative Haptics and Robotics in Medicine (CHARM) Laboratory. The project focused on developing a Unity-based simulation environment for manipulator-equipped soft robotic systems performing shared-autonomy object manipulation tasks.

The simulation framework enabled rapid testing and validation of shared-control algorithms through real-time physics-based interaction modeling, providing a pathway toward future deployment on physical soft robotic platforms. This work received the Best Poster Award at the 2020 UNF SOARS Symposium.