Anylogic and Python-Based Gunshot Detection System Using CNN and Trilateration

Gunshot detection has been applied in fields such as defense and forensic sciences. This plays a very critical role in law enforcement and public safety. Detecting a gunshot begins with automatic detection of the presence of an audio event. These rare events can be analyzed and prevented by using a system that detects the audio and location of gunshots. The success of this system relies ultimately on its accuracy. In this paper, audio data has been collected and a machine learning(ML) model using a Convolution Neural Network (CNN) has been trained that classifies the audio and detects a gunshot acoustic. The prediction of the gunshot location has been implemented using the trilateration method with the use of AI agents as sensors. The model has been trained in Python programming language. This simulation has been implemented in the Java programming language with AnyLogic software as the user interface.