Simulink Student Challenge Winners
MathWorks is happy to announce the winners of the 2020 Simulink Student Challenge. Congratulations to the winners and thank you to all the students who showed off your impressive projects!
Launching a rocket in KSP using Simulink and MATLAB
Northeastern University – Thomas Hayden
This project connects Simulink to the video game Kerbal Space Program (KSP) to control the launch of rocket that delivers a payload to a certain orbit. Thomas uses an API that connects KSP to Java code in a MATLAB Function block inside of the Simulink model. With the sensor data from KSP fed into Simulink, the student's model can perform closed loop control on the flight of the rocket. During the first stage of the launch, the controller tracks the orientation of the rocket and compares this with the desired trajectory as an input to a PID controller. In the second stage, Simulink continues to read in data from KSP and throttles the rocket burn rate to maintain the desired orbit. Overall, this project is a fantastic example of how classical control methods like PID control, coupled with simple modeling practices in Simulink, can be used to create high fidelity models for real-world situations.
ECU programming and HIL test bench development using Simulink
Universidad Politécnica de Madrid – Daniel Garcia
This project explains the development of a torque vectoring control strategy using Simulink and Stateflow. The controller is then validated using generated code from PLC Coder running against the plant model using Speedgoat hardware and Simulink Real-Time. The plant model contains the driver inputs including actuators and the vehicle dynamics using the Vehicle Dynamics Blockset and Powertrain Blockset. The torque vectoring algorithm is implemented using the initial torque, yaw control and traction control using state machines that feeds back into the plant model. Simulink Real-Time Explorer is used to provide actuator inputs and modify driving conditions for real-time testing. Overall, this project showcases an entire workflow of the torque vectoring algorithm's implementation from model to deployment on hardware in an HIL environment entirely using MathWorks' products.
Flying a Thrust Vector Controlled Rocket
ESTACA – Charles Roger
This project demonstrates how Simulink and the Aerospace Blockset can be used to design and simulate the thrust vector control (TVC) system for a model rocket. The control system employs a PID controller to optimize the pitch angle of the rocket in flight via thrust vectoring. The project also makes use of the System Identification Toolbox to estimate a transfer function for the actuator delay in the TVC's servomotors, ultimately reducing the error in the control system. After testing the control system with static fires, the rocket is successfully launched. The video goes on to explain how the sensitivity of the PID controller could be improved to account for model rocket's variable thrust while in flight. Overall, this project provides a great example of how the guidance system for a model rocket can be designed and tested using Simulink.