Investigates how robots can perform more natural and reliable handovers by predicting human intent, motion, and preferred transfer locations. Studies adaptive strategies where the robot reasons about reachability, approach trajectory, grasp affordances, and comfort. Long-term goal is bidirectional handovers — both receiving from and delivering to humans — in a way that feels anticipatory, safe, and socially fluent.

Targets remote teleoperation scenarios where robot lag behind operator intent makes manipulation slow and cognitively demanding. Reduces reaction time by predicting the operator's likely target using eye gaze, hand trajectories, and object context — enabling the robot to proactively move toward the predicted target while keeping the human in control. Makes teleoperation faster, smoother, and less cognitively demanding.

Develops AR-enabled coassembly systems where robots assist humans during tasks involving tools, parts, and shared workspaces. An egocentric AR interface lets the system observe the task from the operator's perspective, track assembly progress, and infer upcoming needs — enabling timely robot assistance (holding parts, fetching tools) without disrupting human workflow. Combines egocentric perception, multimodal sensing, task-state understanding, and safe robot motion.

Develops AR/VR teleoperation interfaces for intuitive robot control via spatially grounded interaction using Meta Quest and handheld gripper interfaces. Operators visualize robot motion, perceive the remote workspace, and provide both high-level and continuous control inputs. Explores how AR, shared autonomy, visual feedback, and predictive robot assistance can reduce operator workload and improve remote manipulation performance.

Investigates how real-time multimodal context improves LLM-based AR task guidance assistants. Develops a context injection framework that progressively incorporates task descriptions, procedural steps, real-time task progress, and hand action data into an LLM's prompt. Validated on the HoloAssist dataset; real-time hand action data provided the largest performance gains in accuracy, relevance, and consistency.

A gamified VR training system for automotive assembly workers combining high-fidelity simulation with interactive game mechanics. Features synchronized multimodal feedback (visual, auditory, haptic) via headset and controllers to study each sensory channel's role in skill acquisition and transfer. Optimized for wireless headsets (Meta Quest 3) with cloud-based analytics dashboards for progress tracking and longitudinal skill assessment. Demonstrated in the context of Kia's Door Module Assembly using a "Ghost Master" scaffold grounded in Cognitive Apprenticeship theory.

Develops a Neuro-Symbolic Immersive Interface integrating LLM-generated narratives into a real-time VR analytics loop to make Graph Neural Network (GNN) decisions interpretable for domain experts. Translates GNN latent representations into natural-language explanations rendered alongside structural graph visualizations. Controlled user studies examine how this multimodal approach affects belief formation, trust calibration, and cognitive load vs. visual-only explanations.

An XR-based teleoperation system developed with the GT Safe Robotics Lab, investigating how haptic, visual, and audio feedback affect operator performance in collaborative manipulation tasks. Seeks to identify the minimum sensory feedback needed to improve teleoperator performance. Uses Meta Quest hand tracking, bHaptics TactGlove vibrotactile haptics, and Unitree G1 humanoid robots with stereoscopic cameras for contact-rich manipulation.

A modular, real-time, gaze-contingent XR framework simulating central vision loss on the Varjo XR-4. Models seven distinct visual phenomena (e.g. spatial distortion, perceptual fill-in) associated with conditions like AMD. Each module is independently configurable per eye. Used for empathy training, inclusive design, and rehabilitation research. Collaboration with Emory University and CVI Atlanta.

AR-guided psychomotor tasks using HoloLens 2 for precision inspection with 3D-printed parts and industrial gauges. Two-session study covering gauge selection, calibration, measurement of part features, and accept/reject decisions based on tolerances.