Currently, there are few places in the UK and Europe that you can test your docking equipment. This means the development of such equipment is slow and expensive, and potentially a barrier to entry for researchers and students. STORM aims to be a facility to validate flight hardware and software prior to launch.
Our project is split into two main halves: the STORM facility and a representative stakeholder. The stakeholder could be a company, academic researcher, student, or anyone who wants to test their docking hardware/software. Our stakeholder consists of a previous master’s project which designed and built a CubeSat docking plate for both the target and chaser satellite. We will use this hardware in STORM as a test case, alongside an RS procured Intel Depth camera for computer vision.
This is a CAD mock-up of the STORM system:
About us and our work
Each member of our team of 6 aerospace engineers are studying the final year of an MEng course at the University of Southampton. As part of this project, we have had the opportunity to visit one of the few satellite docking and rendezvous test facilities in the UK: the IOSM Yard in the Wescott Venture Park, run by the Satellite Applications Catapult.
Meet the team!
Alessandro Borghese – “My role in the STORM project is to design and assemble a 3-axis gimbal, to provide the end effector with roll, pitch and yaw. The gimbal will be lightweight and compact, with aluminium structures holding the motors, and timing belts and pulleys providing the gear reduction necessary to easily rotate the parts.
Below are the roll (left) and pitch (right) mechanisms. The pulley systems (red and green) will give the motors the required torque while also reducing the load-to-rotor inertia ratio and decreasing the degrees per step”.
Dominic Soltau – “My role in the development of STORM is the design, assembly and manufacturing of the hardware, specifically the gantry. This gantry controls the x, y and z motions of the end effector, and houses the 3-axis gimbal, target satellite and controller electronics. This has required several iterations of design to ensure the gantry meets the system requirements and remains below the project budget.
When manufacturing this gantry, several metalwork techniques have been employed such as the lathe, mill, waterjet cutting and a range of metrology equipment to ensure its geometry is correct. Below is a CAD rendering of the 3-axis gantry (left) and the assembled gantry in our testing area (right).”
Jonathan Teo – "I am responsible for the computer vision aspect of this project and am extremely grateful to have RS Components as our sponsor, providing the Intel D435I Depth Camera (202-6352) as shown below. This advanced equipment has significantly improved the accuracy of docking by enhancing the precision of the docking plate's alignment to the target. The RealSense depth camera allows us to capture detailed spatial data, which is crucial for fine-tuning the docking mechanism and ensuring reliable performance. Its high-resolution depth perception and robust technology play a key role in achieving our project goals."
Top image shows the computer vision software in action, computing the distance and orientation of the aruco markers. The lower image shows the Intel D435I Depth Camera sponsored by RS Components.
Roman Bloch – “My work package consists of creating the control algorithm to convert the dynamics outputs into motor control outputs. This is conducted via direct hardware manipulation of the motors. I have also created a graphical user interface (GUI) so we can interact with the robot and run simulations with ease while visualising the outputs in real-time. A preliminary version of the GUI is shown below.”
Rowan Thomas – “I have been working on modelling satellite relative motion for the simulator dynamic model. I am currently validating the trajectories it outputs and will go on to combine it with attitude motion and later, develop guidance and control algorithms to act as an example stakeholder.”
Sam Hinchliffe – “My role in the team so far has been working closely with Rowan, focusing on the rotational dynamics surrounding the simulator dynamic model. Going forwards I will be continuing to validate the rotational motion model, but primarily creating a digital twin of the test rig system itself. A digital twin has many benefits, allowing us to test control inputs and check that the system will behave as expected before risking a physical system test. As well as the test rig digital twin, I am in the early stages of developing a dynamic model (shown below in a still of an early animation) that will allow stakeholders to see what their satellite would look like performing the actuations being tested in our system, if it were in space chasing a real satellite.”
RS Components Impact
A huge thank you to RS for the support with this project. The RS Student fund has allowed us to procure vital fixtures and fittings for the robot, along with the Intel RealSense Depth camera. This camera has elevated the ‘stakeholder’ aspect of our project as it is a much more representative camera that would be used in space compared to our current stakeholder of a Raspberry Pi camera. The support is much appreciated from the whole team.
Conclusions and future work
By May 2025, we aim to have a functional satellite docking test facility. We aim to have validated its performance in a range of scenarios of varying stability and approach corridor using the Qualisys Motion Capture system. This will allow us to quantify the accuracy of STORM to sub-millimetre level, where we can compare the physical output to simulations.
We are happy to answer any questions on our team’s Instagram page, @storm_uos.
Comments