https://drive.google.com/open?id=1ZCC7S7N-Kdg2UQyv4D69ueCIvrC3EWa7wyq9Sc41tQk
https://drive.google.com/file/d/1Idfodw9CoXw1oGwzg6ONh-fTpsPH6W1v/view?usp=sharing
Subject: ai powered, multifunctional, component based, open sourced drone design.
Keywords: ai, open source, drone, weapon, image processing.
State of the art: Currently, drone technology focuses on military and fun usages. There are infinite personal developments that can carry drone technology further. What we mention by saying fun is drone racing. On the other hand, for the military usage drones are using for detection and recognition. After new inventions drones now can eliminate enemies’ weapons. Nevertheless, in security or logistic sides there are not much improvements around. However, open source drone development is supported by many developers. After all, these contributions will be the fundamentals of next level drone technology.
Aim: To design open source drone platform that allow customers to create their own designs for different purposes including military, daily usage etc. without any doubt about data privacy.
Scope: There will be different types of drones and components. These components are available to use on different drones. Their goals are security, personal usage, military usage and logistic usage. Components help to improve features of drones. They are specializations for specific purposes.
Methodology: We made research on drone technologies. After investigate archieves, videos or succesfull-failure experiments we started to design our first prototype. First prototype was only focused on personal usage. Due to the open source environment we were able to find innovative updates and upgrades. We checked the availability of these updates on our first prototype. What we recognized was our prototype couldn’t handle them. We made technical improvements such as expand bottom surface, add new socket plugs for components and fortified engine power. In software phase we used AI technology and open sourced face and voice detection models. Owing to 360 vision cameras we categorized new types of drones.
The main goal of this project is to provide a development platform on which people can develop their own designs for different purposes. Eventual objection of the project is to increase people's living standards by allowing them to design their personal aerial robots for their personal or enterprise purposes.
We will provide a modular architecture design of drone framework for our customers and provide them software development interfaces that they can use for their different purposes.
To solve potential security problems related with drones, we will publish all of the code repositories as open source projects. People will have opportunity to check implemented algorithms and they will have opportunity to fix potential security vulnerabilities. Our main expectation is to get attention of open source community and tech enthusiasts.
Currently, carbon fiber is the lightest and the strongest material in today’s technology. That’s why we selected carbon fiber as main material of our drone design. Carbon fiber is not the most cost effective solution, however durability and robustness are our main focuses. The propellers must be a flexible material. Also, propellers will be modular, it will be easy to replace new propellers and propellers will be really cost effective. Drones will be compatible with machine learning algorithms, all of the analyzed data will be stored in our database. Therefore, movement abilities of the drones will be improved everyday.
To accomplish our goals
We will make agreements with global technology companies to publish their ai products as open source code repositories. At the end, companies will have opportunity to test their complex software products with millions of people and people will access code bases and new technologies which are provided for free. We will make agreements with security forces to build a criminal database. All of the data that generated by drones will be inserted in criminal database. However, the database that we will build will be only accessible by army and security spec. drones.
Professionals are working on unmanned aerial vehicles for about 15 years now. We started playing around with these vehicles in 2005. You can see the exponential growth in the usage of these vehicles. In 1980’s we did not have commercial GPS, therefore it was really hard to develop autonomous flying robots (Nesta).
In 2010, people said that this is going to be a $10 billions industry and people projected all kinds of uses. But it was primarily military uses. It was about surveillance, spying, force protection and warfare. All the kinds of things that many of us are not really interested in. Certainly, from a scientific standpoint, these do not present the challenges or potential for impact (Meola A, 2017).
The Federal Aviation Administration (FAA) predicted that we will have 15,000 drones by 2020. Fast forward for couple of years, and now they are all over the place and 15,000 drones were sold in a single month last year. It is estimated over a million drones were sold just in the Christmas season. It is a $15 billion industry already and again people make famous predictions that it is going to grow to around $20 billion, $25 billion by 2020. You should know this number is also going to be wrong (The Economist, 2015).
What is exciting now is that the applications have grown to areas that we never imagined. Agriculture, border patrols, photography and construction are some of them. So it is really grown to be a very exciting field. Different people call them different things, we prefer to use the word aerial robots, because robots we think of as being smart, they are able to make their own decisions. The military calls them remotely piloted vehicles, because they are not drones. There are human beings that are controlling these vehicles.
The evolution of aerial robots are just starting.
We mostly focus on 5 S’s of aerial robots in this project. The drones that we will developed will have these features, Small If you want to navigate an environment of humans, you want to be small, you want to be able to maneuver. You want to go through doorways. So we are looking for making them small. Safe Clearly, we do not want something banging into humans causing harm, therefore we are trying to make them as safe as possible. Smart If you are building a robot you want it to be smart. Speed We also want these robots to move quickly. Swarms We ask them for moving as a colony like birds.
By dint of these concepts, we started to gave a birth to our first design personal specialization drones. Step by step we added 5S functions on it. The size of the supposed to be small, so we make it smaller than the drones around us. The safety was the sensitive and crucial keyword for us. Therefore, we put some limits and restrictions in our drones. The drone that we created must be smart due to the work properly. If it is not smart then it means it is not eligible for it’s work. The speed was the fourth keyword and many thought that “is it that much necessary?”. However, the speed means time and time means life. If we waste people’s time we wasted their lives too. Time is flying never to return. So, we are still looking, researching to make our drones faster than before. The last concept that we implemented is swarm. Swarm was at the first our arbitrary concept. After a while, we recognized that if count of drones increase, we need an algorithm to prevent any crash, accident. That’s why we designed swarm tactic. When incomprehensible or insoluble happens, drones are available to behave like a swarm until it solves. Additionally, this concept is advantageous for an operational objects.
The 5S is foundation of our project. By these concepts we categorized our drones to the first step which we called frames. The frames are our main category and there are two main frames just for now. They have different purposes and different sub-categorized drone types.
Light Frame: The light frame category is the most important frame for us due to the important duties and of course, the police drones which we called before ruler drones are here. The first prototype that we created was personal specialization. The originality of this drone is, it is common, open-sourced, available for everyone to use, compatible with ordinary updates and very useful for drone developers. As an example of update, there is a compatible lite camera that can put on it. The second type drone of this category is enterprise specialization drones. We already know that we need something to do our work when we are working under hard conditions. For example architectures need laser meter measurement, however, it is not easy to measure some point on skyscrapper. Our drones can do. Also enterprise drones have interface to develop softwares for specific sectors. The last but not least is police drones. Our ruler drones have almost infinite features. They have face detection system, access to criminal database, night vision, bullet dodge, improved camera and even carry a gun. Of course the most important feature is, they have a control on every drones except military drones in heavy frame category. By these features, people can be positive about that our drones will decrease crime incidents in the cities that they live in.
Heavy Frame: The heavy frame category includes more powerful drones. The place that they will be in needs very hard work to do. That’s why the surface design, engine design and of course software is really different from light frame drones. That’s why we seperated them from light frames. The first one is military specialization drones. The basic purpose of these drones are eliminate enemies without any casualities. These drones can into the frame easily and with bullet dodge technology they become very deadly weapons. The second heavy type drone serves to logistic services. They carry heavy and huge stuffs or maybe an important deliveries such as gold or diamonds.
Components: We also developed components which are available to set on drones. Not on all drones but there are variety components that can be suitable with different drone types. For instance, night vision, radar, improved camera, light weapon are available for police drones. For our personal drones lite camera is available just for now. However, as we implemented before that this project is open-sourced project. So, it means for the personal specialization drones developers can setup any software that they want in some restrictions. Weapons, harmful devices etc. are not compatible with our personal drones. Event if they find a way to setup on it, our police drones will prevent it through the control algorithm. For the enterprise drone types we listen our business partners and help them to code the software that they request. On the other hand, for the heavy military drones there are infinite features such as night vision, radar, heavy weapons, improved camera, bullet dodge technology, fortified surface protection, enemy-ally detection, self-destruction etc. For the last type, logistic specialization drones, we keep on fortified surface protection and improved engine power components if it needed.
The drone technology keeps developing day by day since the first prototypes invented by the pioneers of this technology. Owing to be a open-source technology, drones become a personal projects for a developers and designers. Not just in software way, but also in draft way. Admittedly, solo performance included projects fell behind when we compare to first times. Because this technology needs teamwork, idea-sharing and many skills that it is not possible to find in just one person. When the word originality becomes in reality, it is really tough to unearth it. Therefore, brainstorming in professional settled teams are indispensable to carry this technology further. Becoming a professional team is not just mean to introduce impressive, innovative product. As being in this industry, we tried many methods, ways to invent effective drones. After a plenty of time, we brought into the open our big project that includes many varieties to make people’s lives easier, faster, better through the service that we provided. The services that we provided focuses on different areas such as military, security, personal etc. Besides these different service options, our abstract but most important duty is protecting the whole humanity. That’s why, we thought that we need a ruler drone type above all. That’s why we gave a birth to our police drones. For the time being, we completed our all types. The types that we created are much more different than today’s common drones even personal developed drones. We can cavalierly say that behind of the originality of our drones is research and imagination.
We can categorize the methodology under two headings. Hardware and software. These two designs are close-knit each other. For every different type of drones, we constructed different designs.
Hardware Design: The first prototype was not that light to be available for 5S rule. We used iron-copper compound but didn’t get good result. Additionaly, this surface compound had no protection against to electronic waves. When we think 5S’s safe step this common compound won’t ensure safe environment for people. So, we searched different compound and ordered them with outsourcing. In the end, carbon fiber compound fits perfectly. So, we shaped it for different types of drones and also created heavier bulletproof shield component for additional features. The surface of our drones need spots to setup components such as lite camera, guns etc. That’s why we made all of them eligible for any setup. All in all, it took too much time but it is available for usage. We are still trying to find better, faster, innovative ways.
Software Design: The software design is the hardest part of course. Not just only orientation, direction softwares, but also components’ software. The key point here is, this project is open-sourced project. So, we had a lot of help from open-source community. AI-based technologies, entegration, camera and filtering softwares, radar technologies, bullet dodge technology, gun-based fire developments, softwares for mechanical issues such as engine power and of course creating a big, complex database to reach every drone in any urgent situation for police drones. These softwares were really hard to handle and all has to be up-to-date every time. As time progresses, we will keep on checking bugs and errors to fix all thanks to open-source developers.
Work-Time Table Case ID Description of the Work By Whom Date 1 Creating Website for Dronation Undergraduate Students 01/06/2018 2 Creating Visuals for Presentation Undergraduate Students 03/06/2018 3 Drawing Posters Undergraduate Students 04/06/2018 4 Writing Academic Paper Undergraduate Students 06/06/2018
Nesta. Drones: a history of flying robots. https://www.nesta.org.uk/archive-pages/drones-a-history-of-flying-robots/
Meola A (2017). Drone market shows positive outlook with strong industry growth and trends. http://www.businessinsider.com/drone-industry-analysis-market-trends-growth-forecasts-2017-7
The Economist (2015). Welcome to the Drone Age. https://www.economist.com/science-and-technology/2015/09/26/welcome-to-the-drone-age