MSc student in Control Systems @ Iran University of Science and Technology (IUST)
Nonlinear & adaptive control · Multi-agent systems · Swarm robotics · Safety-critical control
I work at the intersection of control theory and robotics — mostly on systems that have to stay stable, safe, and well-behaved when the model is uncertain, the actuators saturate, and the agents have to talk to each other.
- 🎓 MSc in Control Systems at IUST, advised by Dr. Farrokhi
- 🔬 Thesis: Intelligent adaptive formation control for swarm wheeled mobile robots with guaranteed safety and actuator constraints — a decentralized adaptive formation controller in which collision avoidance is a theorem under bounded inputs, built on a braking-distance control barrier function
- 🤖 Long-standing interests: swarm robotics, mobile robots, SLAM, reinforcement learning, and nonlinear/adaptive control theory
- 👨🏫 Instructor of the Digital Control Systems Laboratory; Teaching Assistant in Introduction to Programming, Industrial Networks, Digital Control Systems, and Mechatronics
- 💻 Computer Engineering minor — I like control theory with a working implementation attached to it
- Safety-critical formation control — a relative-degree-one braking-distance control barrier function: with bounded inputs every agent has a finite stopping distance, so collision avoidance becomes a theorem rather than a tuning outcome. Each agent enforces its own half of each pairwise constraint, and inflating the stopping distance by one hold certifies the controller under sampled-data (zero-order-hold) control (paper in preparation)
- SwarmSim — growing a general-purpose MATLAB swarm simulator into a broader research platform for adaptive, learning-based, and distributed control
- Bilateral teleoperation with RBF neural networks — Lyapunov-based adaptive control under communication delay, running on Simulink Real-Time hardware (arXiv preprint)
- Ghaemifar, Goshtasbi, Hajizadeh, Attarzadeh, Riazati, Adaptive RBFNN Control of Uncertain Bilateral Teleoperation Systems with Delay-Dependent LMI Stability Conditions, arXiv:2608.20182, 2026 (under review)
- A paper on braking-distance control barrier functions for input-constrained adaptive formation control (in preparation)
| Repository | What it is |
|---|---|
| SwarmSim | A general-purpose swarm robotics simulation platform in MATLAB, built on a layered architecture so controllers, behaviors, and scenarios can be swapped independently. |
| GNE-Formation | Distributed generalized Nash equilibrium seeking for multi-robot formations with shared collision constraints, under partial-decision information, with a data-driven variant (Python). Self-study implementation of established methods. |
| Vision-Obstacle-Avoidance-Robot | BSc final project: a differential-drive robot that drives to a goal and gets around obstacles using a camera, three ultrasonic sensors and wheel odometry (Raspberry Pi + Arduino, Python/C++). |
| Non-Linear-Control-CourseProject | Feedback linearization vs. sliding-mode control on a single-link flexible-joint arm — symbolically verified models plus a robustness study (parameter mismatch, matched disturbance, chattering). |
| Temperature-IoT-Data-Anomaly-Detection | IoT capstone: anomaly detection on streaming temperature sensor data. |
| Function-Approximation-using-Neural-Networks | MLP-based function approximation — the groundwork behind the neural adaptive controllers above. |
| Area | Tools & Methods | |
|---|---|---|
| 🎛 | Control Theory | Lyapunov stability · Adaptive control · Sliding-mode control · Feedback linearization · Robust & nonlinear control |
| 🛡 | Safety-Critical Control | Control barrier functions (CBF / HOCBF) · QP-based safety filters · Actuator saturation & constraint handling |
| 🤝 | Multi-Agent Systems | Consensus · Formation control · Distributed & decentralized control · Event-triggered and sampled-data control |
| 📐 | Simulation & Modeling | MATLAB · Simulink · Stateflow · Symbolic verification · Monte-Carlo & robustness studies |
| ⚙️ | Real-Time & Embedded | Simulink Real-Time · Hardware-in-the-Loop (HIL) testing · C/C++ · Microcontroller firmware |
| 🧠 | Learning-Based Methods | Neural adaptive control (RBF networks) · Reinforcement learning · CNNs / MLPs · Fuzzy inference systems |
| 🤖 | Robotics | Swarm & mobile robots · Path planning · Manipulator dynamics · Teleoperation |
| 💻 | Software | Python · Git · VS Code · LaTeX · Linux |
Open to collaboration and discussion on multi-agent control, safety-critical control, and robotics research.
- 💼 LinkedIn: Arshia Goshtasbi
- 🎓 Google Scholar: Arshia Goshtasbi
- 📧 Email: Arshia.goshtasbi@gmail.com
"A controller is only as good as the guarantee behind it."