Autonomous Sentry Gun

Abstract

In the past few years image processing and detection algorithms have evolved and developed exponentially and have started to do things that were once believed to be fiction. It has been developed to the point where it is able to outperform the human eye and the time it takes to process things. The most interesting part is that with this fast processing, it also has high accuracy. It is now being trusted to do jobs that were once only assigned to humans. Such applications where image processing and detection algorithms are being used, but not limited to medical surgeries, military technology, security services and food defect detection. The particular field that the team will be looking at in the project and in this report will be regarding the military use and the security services that can be utilized via Digital Image Processing along with Robotics. The team will also be looking at the implications and problems that arise with using such technology in the specified field, and how the team will cater to them. The main reason why such detection systems are not being fully implemented into military use or security services is that the system lacks conscience and still has room for error. These two points are the reason why the team do not see image processing is being involved in weapons, and they continue to use old manually controlled weapons. First and foremost, it is a very widely known fact that the technology that the team have whether it be AI or any sort of General Use Robot, all lack conscience and will act according to the instructions given. This can become an issue if you give complete control to an artificial intelligence or robot. They may mistakenly take someone innocent for a target and attack them. If conscience could be added to such a system, this could be prevented, however, the team lacks the technology to be able to do that. Moreover, there is always room for error. Even if the detection algorithms and image processing are able to detect with 99 percent accuracy, there is still a 1 percent chance of failure. This is significant enough to be fatal. It is also impossible to consider something 100 percent bug-free or accurate when it comes to software engineering. With this in mind, such image-processing and detection-based weapons are not being deployed. Considering the two main problems that has been discussed, the project team plans to develop and design a device that caters to the above concerns, while also ensuring safety via the device. The team will be using Digital Image Processing and Robotics to achieve the mentioned system and constrict it such that it satisfies the above-mentioned flaws