Ardalan Aryashad

I’m a third-year AME Ph.D. student at USC, advised by Prof. Yan Jin. I have recently started an M.S. in Computer Science. At the Impact Lab, I am currently working on vision-language-action (VLA) models and their applications in robotic task planning and manipulation, and agentic AI approaches for MBSE modeler tasks.

Email  /  CV  /  Scholar  /  LinkedIn  /  Github

profile photo

Publications

VLAG icon
VLAG: Graph-Based Planning for Vision-Language-Action Models
IDETC/CIE 2025

Modular graph-based planner for long-horizon manipulation with dedicated vision, language, and action modules.

Fog Removal Benchmark icon
Fog Removal Benchmark
WACV Workshop

Benchmarking defogging methods by downstream task performance on synthetic and real datasets.

Selected Projects

SEMAA logo
SEMAA: AI Agent Assistant for MBSE
USC IMPACT Laboratory

AI agent assistant for MBSE workflows, featuring a context-aware chat interface, diagram import/export, and a web-based plugin to speed up modeling tasks.

USC Racing telemetry dashboard
USC Racing: Suspension and Data Acquisition
USC Racing Team

Combined suspension and telemetry projects focused on quarter-car modeling, genetic algorithm tuning, and a data pipeline with dashboards for performance analysis.

Biped robot walking demonstration
A Walking Biped Robot
AME 556: Robotics
Fall, 2024

A 2D biped robot model using MATLAB Simscape with MPC control for stable locomotion, including walking forward, backward, and stair climbing capabilities.

Automotive predictive maintenance application dashboard
Automotive Predictive Maintenance Application
AME 505: Engineering Information Modeling
Fall, 2023

A machine learning application that uses real car sensor data to predict engine problems. Features data preprocessing, multiple ML models, and a mobile-friendly dashboard built with Flet.


Feel free to use this website's source code, design courtesy of Jon Barron.