Zhishan Wang

I lead and contribute to the design of a closed-loop autonomous racing control framework that integrates a vehicle dynamics model, controller implementation, and MATLAB-based evaluation. The project began with a baseline PID design and gradually moved toward ADRC-based and optimization-aware approaches to better handle disturbance rejection and racing constraints.

• Built a physics-based simulation pipeline around planar bicycle dynamics, tire behavior, longitudinal resistance, actuator feasibility, and repeatable closed-loop testing.
• Designed and tuned PID controllers as an interpretable baseline, then explored ADRC variants to improve robustness, smoothness, and speed–safety trade-offs.
• Evaluated controllers using multi-objective criteria including lap time, safety violations, and steering smoothness, with results showing faster lap performance and lower violation rates than the baseline.
• Presented this work at the 2026 CCAT Global Symposium Student Poster Competition , where it was recognized as the Undergraduate Winner; the project was also honored with the ITS Michigan Scholar Award for an outstanding undergraduate-level entry.

This semester-long journal records the transition of my work from single-vehicle controller benchmarking to a broader autonomous racing framework involving multi-vehicle interaction, multi-agent coordination, resilience analysis, and strategy-level decision design.

• Outlined a roadmap connecting strategy, estimation, low-level control, robustness verification, and stability analysis.
• Formulated multi-vehicle racing with explicit track constraints, collision-avoidance sets, and coordination variables.
• Connected these models to PID and MPC-style ideas, with attention to safety filtering and distributed information patterns.

This project studies soccer team formation coordination from a multi-agent control perspective. Instead of treating team shape as a fixed set of absolute positions, I model formation as a task-dependent relational structure, where players preserve tactically meaningful distances and role-based constraints under local information.

• Modeled soccer formation as a task-dependent weighted tactical graph rather than a rigid absolute template, allowing the team structure to adapt to tactical tasks such as defending, pressing, and transition.
• Proposed a distributed estimation-control framework in which each player estimates a latent team state from local observations, updates a task-dependent tactical graph, and performs projected local control under role feasibility constraints.
• Designed a small five-player defensive-block simulation to illustrate defensive compactness recovery, fullback-pressing reconfiguration, and fast agreement in distributed estimation.
• This project connects my long-term interest in football with ideas from multi-agent autonomy, distributed estimation, graph theory, and constrained control.

In this research experience, I studied how real-time feedback and optimization can be used to improve regulation and redistribution in energy systems. The work combines system understanding, modeling, and optimization-based analysis under uncertainty.

• Analyzed energy system operation while considering weather and user preference.
• Used MATLAB, Simulink, and CVX to build optimization-driven simulation workflows.
• Studied performance over time and the relationship between marginal benefit and output capacity.

I designed and implemented a photoplethysmography-based optical heart rate sensor capable of measuring heart rate in the 40–200 BPM range. This project strengthened my understanding of analog signal chains, noise suppression, and circuit-level system integration.

• Selected and integrated infrared sensing components, driving circuits, and filtering stages.
• Balanced gain distribution to avoid saturation while preserving robust heart rate detection.
• Refined filter parameters and hysteresis behavior to address low signal amplitude and noise.

For this FPGA-based design project, I built a lunar lander control system that handled speed, altitude, and spacecraft state logic through hardware description and sequential design.

• Developed a digital design workflow using SystemVerilog and state-machine thinking.
• Constructed the physical and logical structure of the controller and validated it against required metrics.
• This project received an A+ and helped me bridge mathematical reasoning and hardware execution.

This earlier research experience gave me hands-on exposure to modeling, structural analysis, and feedback-oriented thinking in an electromechanical setting.

• Used COMSOL and SolidWorks to study structural feasibility and support system design.
• Contributed to piezoelectric material selection and testing.
• Designed feedback-related reasoning based on electrical output and response behavior.