Empowering Object Detection in Dynamic Environments with AI-Driven MIMO Radar Technology

Millimeter-wave radars are increasingly used for surveillance and automotive applications, as optical-based systems have a hard time in complex environments such as night, fog, etc. These applications use advanced object detection and classification algorithms that require huge amounts of datasets for training and testing, which is hard to obtain considering that they utilize the same radar configuration and contain required scenarios that enhance your model accuracy. Moreover, in signal processing perspective, angular resolution is constrained by the number of receivers present in the radar. So, to account for these problems, we propose a virtual radar that can generate datasets through code, which can be used to train and test a model, and an AI model that won't be constrained by angular resolution.