SLAAM (Student Learning and Achievement Aerospace and Mechanical)

This project introduces students to how generative artificial intelligence can be used for engineering design. Students will learn how image-based generative AI models can be adapted to create structural geometries that satisfy performance targets such as prescribed deformation, stiffness, and stress limits. They will run guided experiments where they generate candidate designs and evaluate them using engineering simulations. The project exposes students to modern computational engineering workflows that combine physics-based reasoning and AI methods. Basic familiarity with Python is helpful but not required. 

Students will explore airflow around a missile traveling at hypersonic speeds. Starting from a 3-D computer-aided design (CAD) model, students will learn how engineers generate computational meshes and run physics-based simulations to compute the flow field around the missile body. We will begin with a simplified (inviscid) simulation and then move onto a more realistic case that includes air friction effects. Students will visualize shock waves, Mach number and pressure distributions using professional aerospace engineering tools. This project introduces the complete engineering workflow used in real-world aerospace research, both in industry and academic settings - from geometry creation to post-processing and analysis. 

Students will use multi-physics simulations in the software package COMSOL to solve the coupled heat transfer and chemical kinetics of thermal runaway in Li-ion battery cells. Specifically, the students will learn how to set up a transient heat transfer simulation, how to define spatiotemporal functions of heat generation rate within the battery cells, and how to simulate the thermal runaway process in a single cell. Next, students will extend their single-cell model to a multi-cell battery module and simulate the process of cascading thermal runaway and propose methods to suppress the cascading effect. 

This project focuses on the development and physical implementation of advanced offensive and defensive maneuvers for an Autonomous Foosball Table (AFT). While previous iterations relied heavily on 2D physics simulations to inform logic, this program will challenge students to program and execute strategies directly on the physical hardware. Students will leverage a system utilizing stepper and servo motors for lateral and rotational rod control, integrated with real-time computer vision for ball tracking. The core objective is to move beyond basic "kick and run" behaviors to a sophisticated, physically realized playbook including: Offensive Execution - Developing C++ and Python-based control logic for complex shots such as the snake shot, push/pull shots, and coordinated lateral passing, Dynamic Defense - Programming reactive blocking algorithms that utilize predictive ball trajectory data to position figures effectively, Strategic Integration - Mapping out a high-level decision-making hierarchy that determines when to trap, pass, or shoot based on live field conditions. By focusing on physical trials, students will address real-world robotics challenges such as motor latency, mechanical bottlenecks, and environmental noise in computer vision. This project provides a tangible testbed for mastering multi-agent coordination and high-speed robotic manipulation.