MIMO controller for control of Satellite Constellation (MATLAB/SIMULINK)
- or -
Post a project like this$1.8k
- Posted:
- Proposals: 7
- Remote
- #2266944
- Expired
Artificial Intelligence| machine Learning |Deep Learning |Computer Vision |Medical Imaging | NLP
Kimhae
wordpress, iOS development, app development , logo,PHP Developer - Drupal, Wordpress,Codeigniter,Zend Framework, Magento,Openc
London
Prototyping on embedded systems, PCB, Graphic Designer, Short Ad videos, Video Editing, Logo Designing
Alleppey
2165633179906023066542399451251956325198442546246
Description
Experience Level: Expert
I would like for a fuzzy sliding mode controller to be developed that will position three satellites in a linear constellation and then align broadside to a stationary object for transmitting and receiving capabilities. The fuzzy sliding mode controller in this thesis paper [p.30] I believe will work (https://pdfs.semanticscholar.org/da49/96a8052608343c08961deec16882349ebe61.pdf) at the bottom of the paper a source code is given so it can be modified and tested.
The actuators I would prefer to use are 3 magnetorquers per satellite. This thesis paper has modeled 3 magnetorquers, 1 for each x,y,z axis (https://digitalcommons.mtu.edu/cgi/viewcontent.cgi?referer=&httpsredir=1&article=1895&context=etds) again at the bottom is source code to recreate this. The inertia matrix needs to be changed to a CubeSat.
This paper contains the type of satellite I would like to be modelled and shows the sensor and actuator modeling. Refer to Table 1: Magnetorquer Parameters to determine the max control torque that can be generated by the magnetorquers.(https://www.researchgate.net/publication/258396753_Design_of_Attitude_Control_Systems_for_CubeSat-Class_Nanosatellite) and includes on [p.12] the simulation information with inertia matrix = [.002 .002 .002]kgm2. This paper includes a model for the magnet sensor, magnetometer, which needs to be modeled in the plant of the system.
The position estimates of the CubeSat will be provided by a StarTracker. This paper shows how the StarTracker can generate quaternion coordinates, and it shows how the extended Kalman filter (EKF) can remove the noise from the quaternion coordinates.(https://www.researchgate.net/publication/307877953_Spacecraft_Attitude_Estimation_Based_on_Star_Tracker_and_Gyroscope_Sensors) I would like to have a plot showing the quaternion coordinates for all three satellites with noise generated by the StarTracker device and another plot showing the extended kalman filter successfully removing the noise so the quaternions can be used as accurate attitude information.
Please see this for a matlab script of an extended kalman filter (http://www.blogdugas.net/blog/2015/05/10/extended-kalman-filter-with-quaternions-for-attitude-estimation/)
Please see this for matlab examples of satellite orbits modeled (https://www.mathworks.com/matlabcentral/fileexchange/55533-satellite-orbits-models-methods-and-applications)
Please see this for matlab script of determining position, velocity, and acceleration between the satellites (https://smallsats.org/2013/05/16/relative-motion-of-satellites-numerical-simulation/) I think this will be useful to check if they are aligned linearly, i.e. the velocity's and acceleration will be the same and the position will be aligned on one of the axis in order to be linear.
please see the following links for the CubeSat parts talked about in the papers.
https://www.cubesatshop.com/product/nctr-m002-magnetorquer-rod/
https://www.cubesatshop.com/product/1-unit-cubesat-structure/
**if for some reason the magnetorquer can not provide enough torque to accurately align the satellites try implementing the three magnetorquers and a thruster https://www.cubesatshop.com/product/ifm-nano-thruster/
I have attached my proposal for the design. I would like a simulink model for the overall system and plots that prove the satellites are aligned in a linear fashion and broadside to a stationary object while meeting the requirements of not exceeding control power available. The settling time is not of real importance, this is mainly a power and accuracy constraint. The key plots are the quaternion coordinates of all three satellites which will show their constellation becoming linearized and stabilized, this will yield them all having the same velocity and acceleration once they are in proper formation, and the control input that the sliding fuzzy controller generates being a feasible torque value that the magnetorquer explained above can provide. Please document any matlab scripts for easy readability and label all simulink blocks.
I understand that this MIMO controller is good deal of work, so please don't hesitate to ask me for more information or clarify some of the details.
The actuators I would prefer to use are 3 magnetorquers per satellite. This thesis paper has modeled 3 magnetorquers, 1 for each x,y,z axis (https://digitalcommons.mtu.edu/cgi/viewcontent.cgi?referer=&httpsredir=1&article=1895&context=etds) again at the bottom is source code to recreate this. The inertia matrix needs to be changed to a CubeSat.
This paper contains the type of satellite I would like to be modelled and shows the sensor and actuator modeling. Refer to Table 1: Magnetorquer Parameters to determine the max control torque that can be generated by the magnetorquers.(https://www.researchgate.net/publication/258396753_Design_of_Attitude_Control_Systems_for_CubeSat-Class_Nanosatellite) and includes on [p.12] the simulation information with inertia matrix = [.002 .002 .002]kgm2. This paper includes a model for the magnet sensor, magnetometer, which needs to be modeled in the plant of the system.
The position estimates of the CubeSat will be provided by a StarTracker. This paper shows how the StarTracker can generate quaternion coordinates, and it shows how the extended Kalman filter (EKF) can remove the noise from the quaternion coordinates.(https://www.researchgate.net/publication/307877953_Spacecraft_Attitude_Estimation_Based_on_Star_Tracker_and_Gyroscope_Sensors) I would like to have a plot showing the quaternion coordinates for all three satellites with noise generated by the StarTracker device and another plot showing the extended kalman filter successfully removing the noise so the quaternions can be used as accurate attitude information.
Please see this for a matlab script of an extended kalman filter (http://www.blogdugas.net/blog/2015/05/10/extended-kalman-filter-with-quaternions-for-attitude-estimation/)
Please see this for matlab examples of satellite orbits modeled (https://www.mathworks.com/matlabcentral/fileexchange/55533-satellite-orbits-models-methods-and-applications)
Please see this for matlab script of determining position, velocity, and acceleration between the satellites (https://smallsats.org/2013/05/16/relative-motion-of-satellites-numerical-simulation/) I think this will be useful to check if they are aligned linearly, i.e. the velocity's and acceleration will be the same and the position will be aligned on one of the axis in order to be linear.
please see the following links for the CubeSat parts talked about in the papers.
https://www.cubesatshop.com/product/nctr-m002-magnetorquer-rod/
https://www.cubesatshop.com/product/1-unit-cubesat-structure/
**if for some reason the magnetorquer can not provide enough torque to accurately align the satellites try implementing the three magnetorquers and a thruster https://www.cubesatshop.com/product/ifm-nano-thruster/
I have attached my proposal for the design. I would like a simulink model for the overall system and plots that prove the satellites are aligned in a linear fashion and broadside to a stationary object while meeting the requirements of not exceeding control power available. The settling time is not of real importance, this is mainly a power and accuracy constraint. The key plots are the quaternion coordinates of all three satellites which will show their constellation becoming linearized and stabilized, this will yield them all having the same velocity and acceleration once they are in proper formation, and the control input that the sliding fuzzy controller generates being a feasible torque value that the magnetorquer explained above can provide. Please document any matlab scripts for easy readability and label all simulink blocks.
I understand that this MIMO controller is good deal of work, so please don't hesitate to ask me for more information or clarify some of the details.
Michael W.
100% (1)Projects Completed
2
Freelancers worked with
1
Projects awarded
50%
Last project
19 Mar 2019
United States
New Proposal
Login to your account and send a proposal now to get this project.
Log inClarification Board Ask a Question
-
There are no clarification messages.
We collect cookies to enable the proper functioning and security of our website, and to enhance your experience. By clicking on 'Accept All Cookies', you consent to the use of these cookies. You can change your 'Cookies Settings' at any time. For more information, please read ourCookie Policy
Cookie Settings
Accept All Cookies