Overview

The GNC (Guidance, Navigation and Control) team is an Integrated Project Team (IPT) which is part of the Avionics department. The main responsibility of the team is to develop, implement and test control algorithms that allow the rocket to automatically perform actions needed for a successful mission. Our work includes modelling the systems under control, identifying key parameters and using them to design control algorithms. The algorithms are then rigorously tested through Matlab and Simulink simulations as well as hardware-in-the-loop tests. By using sensor fusion techniques to estimate flight trajectories and controlling airbrake deployment and engine shutdown for precise apogee targeting, we apply theoretical and mathematical models to real-world applications, allowing the rocket to operate autonomously!

Guidance

The Guidance subsystem is responsible for determining the reference trajectory for both the rocket and the parafoil, based on the state estimates provided by the Navigation subsystem. This trajectory enables the Control subsystem to apply corrective or control actions to maintain the desired flight path. A critical element for mission success is the actuation of the airbrakes, achieved through interpolation between reference trajectories computed with maximum and minimum opening, depending on the estimated mass and vertical velocity. Similarly, during the guided recovery descent, the Controlled Altitude System for Canopy Adjustments (CASCA) directs the parafoil towards the target by interpolating between curves with maximum and minimum turn rates, selected based on altitude and distance from target location.

Navigation

The Navigation subsystem is dedicated to estimating all the states of the rocket: the Navigation and Attitude System (NAS) is the algorithm responsible for the on-board live estimation of the rocket’s position, velocity and attitude. Thanks to a Multiplicative Extended Kalman Filter, the algorithm fuses together measurement coming from different sensors, such as accelerometers, gyroscopes, barometers, magnetometers, while minimizing the effect of noises and disturbances.

Coupled with that, the Mass Estimation Algorithm (MEA) is dedicated to the estimation of the rocket mass, given the data coming from the engine sensors and a model of the combustion chamber pressure.

All the estimated states are used by the guidance and control subsystems, to make decisions with the most up to date information possible

Control

The Control subsystem interfaces directly with actuations on the rocket: using the estimated states, it commands the rocket along the trajectories computed by the guidance subsystem. To achieve the most accurate apogee possible, it manages critical tasks such as shutting down the engine at the correct time, as well as regulating the airbrakes to adjust aerodynamic drag, according to the chosen trajectory. During the parafoil descent, the control subsystem steers the parafoil wing to reach the target as precisely as possible.

ADA

The Apogee Detection Algorithm (ADA), the rocket’s most critical algorithm, is responsible for reliably identifying when the rocket reaches apogee. Accurate detection enables correct activation of the expulsion mechanism, ensuring safe deployment of the recovery systems. Given its critical role, simplicity and redundancy are key priorities of the algorithm: the ADA employs a majority voting system and relies solely on barometric measurement, augmented by a linear Kalman filter for state estimation. Additionally, the estimated vertical position is used to trigger the deployment of the main parachute at the prescribed altitude.

ARP

The Autonomous Rocket Pointer (ARP) is a device capable of tracking the rocket during all phases of flight. The team working on this subsystem includes members from other IPTs, such as SWD, ELC, GNC, STR, and MMC, and is responsible for the system’s design, development, and testing in all its aspects. The ARP system mounts a set of directional antennas used to retrieve and store live telemetry from the rocket, along with a camera. It uses output from the NAS, received via telemetry, to estimate the rocket’s position and combines that data with its own orientation to automatically compute control actions sent to the motors. The software also implements a propagation algorithm that mitigates the effects of temporary telemetry loss and contributes to smoother motion. A simulation environment has been developed to test and evaluate the system. During testing, it enables the transmission of past flight telemetry, recreating a scenario that closely resembles actual flight conditions, which is useful for assessing the ARP’s performance.

Requirements

Base

  • Willingness to learn and work autonomously
  • Familiarity with linear algebra
  • Basic knowledge of Matlab and Simulink
  • Knowledge of the basics of control theory
  • Sensors and measuring instruments

Advanced

  • Optimal estimator and Kalman filter theory
  • State Space (SS) and Model Based (MPC) Control Design
  • Basic knowledge of C++
  • Advanced Knowledge of Matlab and Simulink (familiarity with most common toolboxes)
  • System Identification