Radar sensors can be used for analyzing the induced frequency shifts due tomicro motions in both range and velocity dimensions identified as micro-Doppler($\boldsymbol{\mu}$-D) and micro-Range ($\boldsymbol{\mu}$-R) respectively.Different moving targets will have unique $\boldsymbol{\mu}$-D and$\boldsymbol{\mu}$-R signatures that can be used for target classification.Such classification can be used in numerous fields such as gait recognition,safety and surveillance. In this paper, a \unit[25]{GHz} FMCW Single InputSingle Output (SISO) radar is used in industrial safety for real-timehuman-robot identification. Due to the real-time constraint, jointRange-Doppler (R-D) maps are directly analyzed for our classification problem.Furthermore, a comparison between the conventional classical learningapproaches with handcrafted extracted features, ensemble classifiers and deeplearning approaches is presented. For ensemble classifiers, a restructuredrange and velocity profiles are passed directly to ensemble trees such asgradient boosting and random forest without feature extraction. Finally, a DeepConvolutional Neural Network (DCNN) is used and raw R-D images are directly fedto the constructed network. DCNN shows a superior performance of 99\% accuracyin identifying humans from robots on a single R-D map.
展开▼
