Search for: All records

Vision-based methods are commonly used in robotic arm activity recognition. These approaches typically rely on line-of-sight (LoS) and raise privacy concerns, particularly in smart home applications. Passive Wi-Fi sensing represents a new paradigm for recognizing human and robotic arm activi- ties, utilizing channel state information (CSI) measurements to identify activities in indoor environments. In this paper, a novel machine learning approach based on discrete wavelet transform and vision transformers for robotic arm activity recognition from CSI measurements in indoor settings is proposed. This method outperforms convolutional neural network (CNN) and long short- term memory (LSTM) models in robotic arm activity recognition, particularly when LoS is obstructed by barriers, without relying on external or internal sensors or visual aids. Experiments are conducted using four different data collection scenarios and four different robotic arm activities. Performance results demonstrate that wavelet transform can significantly enhance the accuracy of visual transformer networks in robotic arms activity recognition. 

This study addresses the challenge of selecting sensors for linear time-varying (LTV) systems dynamically. We present a framework that designs an online sparse sensor schedule with performance guarantees using randomized algorithms for large-scale LTV systems. Our approach calculates each sensor’s contribution at each time in real-time and immediately decides whether to keep or discard the sensor in the schedule, with no possibility of reversal. Additionally, we provide new performance guarantees that approximate the fully-sensed LTV system with a multiplicative approximation factor and an additive one by using a constant average number of active sensors at each time. We demonstrate the validity of our findings through several numerical examples.