Sandbox: a novel algorithm for object recognition using millimeter wave radar

Focusing on the millimeter wave (MMW) radar classification and recognition problem of objects on the road, such as pedestrians, cyclists, and cars, a sandbox algorithm based on a three-dimensional feature vector, including the target motion feature and electromagnetic (EM) scattering feature radar cross section (RCS), is proposed. And then, a new machine learning algorithm suitable for MMW radar target classification is proposed and verified in the self-test data set. Compared with conventional algorithms, such as k-nearest neighbor (KNN) and support vector machine (SVM), it has better classification accuracy and faster classification speed.