Design of autopilot using positioning system and removal of attitude sensor system for unmanned aircraft

Dissertation for receiving a master's degree (M.Sc)

Chapter 1: Overview of the plan

1-1 Statement of the problem

The complete autopilot system is a set of hardware and software that, while controlling the bird, allows viewing the position, speed and status of the bird, as well as crossing waypoints provides Common systems usually include position measurement sensor (AHRS [1]) and position indicator sensor (GPS [2]), the presence of position measurement sensor in this set will significantly increase the price of the whole set. In addition, not being able to access it quickly will increase the time of building the collection. By using available GPS data and a series of mathematical relationships, it is possible to calculate the position of the bird, or in other words its roll angles[3], elevation[4], direction[5]. One of the major bottlenecks of the aerospace industry is to reduce and instead prepare a product that gives it the ability to navigate and guide the bird according to the GPS standards, which has been through the design and modification process for years. This low-cost navigation system [6] can be used as a stabilizer and autopilot system in low-cost drones such as training birds, hand-launched, target, etc. to be used.

1-2 The purpose of designing an autopilot system with GPS

What can be defended as a specific result of this research is to provide a practical and applicable scientific method that can be used to extract navigation information without having a position sensor system. The use of such a method in inexpensive unmanned birds such as target birds [7], training, hand-launched and micro and disposable suicide birds seems to be very important and practical. 

1-3 reasons for the importance of removing position sensor systems

Currently, the available navigation systems are generally GPS/AHRS, the presence of AHRS in it will increase the price of the entire set, which is not economically justified in cheap birds, and due to the fact that most of these sensors are imported and banned, it is not possible to access them quickly, therefore, it will also increase the time to build the set.

Regarding the previous speech, the reasons for this research can be attributed to the following:

Increasing the reliability factor of navigation against the common problems of common attitude sensors (instability and saturation [8] IMU during flight)

Reducing the significant costs of purchasing and building guidance and control systems

Increasing the production of cheap drone products to train troops and carry out test missions

Decreasing the depreciation of the airline fleet

Stopping dependence on foreign countries, in order to procure military parts that are always subject to embargo (due to the use of commercial electronic products that are always available in the domestic market)

1-4 key questions

In the industry, what is of great interest today is the use of general tools to build new equipment, and it is referred to as COTS technology [9].Among the benefits of using these tools, the following can be mentioned:

Usually, a long time has passed since their creation, so many modifications have been applied to them

They are cheaper compared to similar products, due to the production in high circulation

It is always possible to access them

are designed to be user-friendly[10]

On the other hand, these products have defects such as the following:

Lack of sufficient accuracy

Lack of necessary resistance against external physical phenomena

Low reliability

Products produced in the autopilot industry are always designed in the military category, due to the need to create a strong, resistant and reliable fleet, and the cost of making them is extremely high. Although the production of such products is necessary in the air fleet itself, it is not necessary in training sectors, hand-launched birds [11], target birds.

AHRS, as the heart of navigation systems, is one of these systems that is always a problem in the air fleet. In this research, an attempt has been made to answer these questions:

Is it possible to estimate the body position of a bird using the output information of a GPS?

Does the complexity of calculations using the Kalman filter algorithm decrease to the point where this information can be used in guidance and control?

What are the limitations of the new sensor? ?

What is the difference between the calculated information and the output values ??of AHRS?

Also, to advance the research, it is initially accepted that:

The output information does not necessarily correspond exactly to the bird's physical condition, and it is valid to the extent that the general missions of training and which was mentioned earlier.

This autopilot cannot be used alone in sensitive military missions

Substitution of pseudo-elevation and pseudo-aspect angles instead of the required altitude and direction angles for guidance and control in the autopilot

1-5 simulated model

The selected drone is a surveillance bird with a net weight of 3 kg and a wingspan of 2 meters. Therefore, to ensure the correctness of the work method and its accuracy, this bird is used as a test sample. The values ??of roll angles, peak and direction are calculated using GPS data in the simulator and compared with its original values, and finally control loops are applied on it. 1-6 General definitions of variables and keywords ? : Roll angle of flying devices ? : Pitch angle Flying devices

?: Heading angle of flying devices

Figure 1.1 showing the body coordinate system and other parameters on the bird.

xb, yb, zb: axis of the body coordinates of the plane[12]

Vg: GPS velocity vector

?: flight path angle

: pseudo-roll angle

1-7 required information in autopilot

The minimum parameters that should be present in an autopilot system are altitude, speed, length Geography, latitude, roll angle, elevation angle and side angle of the bird. Except for the last three, other parameters can be directly obtained from GPS. The pseudo-side angle [13] provided by GPS can be used as a side angle with a good approximation. The pseudo peak angle of the bird, which can be calculated by mathematical relations and using the global positioning system ground speeds, can be considered as the peak angle with a good approximation. The roll angle can be fully calculated by mathematical relations using the VNED output velocities and accelerations, where the Kalman filter is used to estimate the estimated incoming accelerations using GPS speed measurements. With a certain number of acceptable assumptions and simplifications, it is possible to establish a relationship between the positions and accelerations of the flying device estimated from the GPS speed measurements.

Figure 1.