Micro-doppler based human-robot classification using ensemble and deep learning approaches

Abstract

Radar sensors can be used for analyzing the induced frequency shifts due to micro-motions in both range and velocity dimensions identified as micro-Doppler (µ-D) and micro-Range (µ-R), respectively. Different moving targets will have unique µ-D and µ-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 25 GHz FMCW Single-Input Single-Output (SISO) radar is used in industrial safety for real-time human-robot identification. Due to the real-time constraint, joint Range-Doppler (R-D) maps are directly analyzed for our classification problem. Furthermore, a comparison between the conventional classical learning approaches with handcrafted extracted features, ensemble classifiers and deep learning approaches is presented. For ensemble classifiers, restructured range and velocity profiles are passed directly to ensemble trees, such as gradient boosting and random forest without feature extraction. Finally, a Deep Convolutional Neural Network (DCNN) is used and raw R-D images are directly fed into the constructed network. DCNN shows a superior performance of 99% accuracy in identifying humans from robots on a single R-D map.

Person Identification and Body Mass Index: A Deep Learning-Based Study on Micro-Dopplers

“…For instance, differentiation between human activities such as walking, running and crawling based on their µ-D signatures [9], [10]. Other studies used the µ-D characteristics to differentiate between humans and other moving targets [11], [12].…”

Human Motion Detection through Wall Based on Micro-Doppler Using Kalman Filter Combined with Convolutional Neural Network

“…In this context, each moving object has a unique MD and micro-Range (MR) signature that can be used for classification of this object. [2]. Therefore, in the detection and recognition of human motion, many existing works are presenting different techniques and sensors for human detection and activity recognition, which are different from our work.…”

“…where; Wv -motion frequency in v direction. Then, the Range-Doppler (R-D) map is constructed from the matrix Msp, [2];…”

“…Each one of these sensors has a specific limitation. The monitoring areas such as; suspicious event, sporadic conduct, uncommon behavior, unusual behavior, abnormal behavior, and anomaly [1][2][3][4].…”

“…Therefore, in the detection and recognition of human motion, many existing works are presenting different techniques and sensors for human detection and activity recognition, which are different from our work. In [2] a 25 GHz radar is used, and MD signatures are used with conventional machine learning and Deep Convolutional Neural Networks (DCNN) approach to detect possible human presence. In [9] was used Method of Moments enhanced with Fast Multipole Method (MoM-FMM) with 5GHz radar, this algorithm is used for scattering model.…”