Journal of Ambient Intelligence and Smart Environments

The purpose of this paper is to offer a unique adaptive path planning framework to address a new challenge known as the Unknown environment Persistent Monitoring Problem (PMP). To identify the unknown events’ occurrence location and likelihood, an unmanned ground vehicle (UGV) equipped with a Light Detection and Ranging (LIDAR) and camera is used to record such events in agriculture land. A certain level of detecting capability must be the distinct monitoring priority in order to keep track of them to a certain distance. First, to formulate a model, we developed an event-oriented modelling strategy for unknown environment perception and the effect is enumerated by uncertainty, which takes into account the sensor’s detection capabilities, the detection interval, and monitoring weight. A mobile robot scheme utilizing LIDAR on integrative approach was created and experiments were carried out to solve the high equipment budget of Simultaneous Localization and Mapping (SLAM) for robotic systems. To map an unfamiliar location using the robotic operating system (ROS), the 3D visualization tool for Robot Operating System (RVIZ) was utilized, and GMapping software package was used for SLAM usage. The experimental results suggest that the mobile robot design pattern is viable to produce a high-precision map while lowering the cost of the mobile robot SLAM hardware. From a decision-making standpoint, we built a hybrid algorithm HSAStar (Hybrid SLAM & A Star) algorithm for path planning based on the event oriented modelling, allowing a UGV to continually monitor the perspectives of a path. The simulation results and analyses show that the proposed strategy is feasible and superior. The performance of the proposed hyb SLAM-A Star-APP method provides 34.95%, 27.38%, 33.21% and 29.68% lower execution time, 26.36%, 29.64% and 29.67% lower map duration compared with the existing methods, such as ACO-APF-APP, APFA-APP, GWO-APP and PSO-APP.

Human action recognition (HAR) plays an important role in social interaction in various fields. This study proposes a light-weight skeleton and two-layer bidirectional LSTM-based Seq2Seq model (SB2_Seq2Seq) for HAR to trade off recognition accuracy, users’ privacy and computer resource usage. An experiment was conducted to compare the proposed SB2_Seq2Seq with other skeleton-based Seq2Seq models and non-skeleton RGB video frame-based LSTM, CNN and seq2seq models. The UCF50 dataset was used for model evaluation, where 60%, 20% and 20% for model training, validation and testing, respectively. The experimental results show that the proposed model achieves 93.54% accuracy with 0.0214 Mean Square Error (MSE), suggesting that the proposed model outperforms all the other models. Besides, it also shows that the proposed model achieves state-of-the-art accuracy compared with state-of-the-arts methods in literature.

As the world’s population rises, the healthcare system experiences significant changes. Wireless body area network (WBAN) is an emerging technology that has considerable impact on medical and non-medical applications. However, two crucial challenges in WBANs are interference minimization and channel assignment. High interference may increase collision probability, transmission delay, and energy consumption. Multichannel schemes are proposed to reduce the data transmission latency and improve the system throughput by allowing simultaneous transmission of sensors in coexisting WBANs. When WBAN users move, they need to switch the channels frequently to avoid potential channel conflicts and to maintain the Quality of Service (QoS). However, frequent switching may raise energy consumption and aggravate delay. Existing multichannel assignment schemes failed to perform well in highly dynamic and densely deployed WBANs environments. In contrast to existing studies, this paper proposes a Prediction-based Channel Assignment (PCA) algorithm that selects the channels for WBANs to remain valid for future time instances and thus minimizes the delay and number of channel switches for dynamic and coexisting WBANs. When a WBAN needs to switch a channel, the proposed method predicts the future neighbors of that WBAN based on its history. It explores the channel information of present and future neighbors to select a suitable channel with higher resilience in a dynamic environment. Thus, our algorithm minimizes channel interference by avoiding unnecessary channel switching. We have used machine learning algorithms to predict the future neighbors of a WBAN. Experiment results show that the proposed algorithm performs better than an existing algorithm and random channel assignment in delay and throughput.

Mobile Ad Hoc Networks (MANETs) are self-organizing, self-configuring, and infrastructure-less networks for performing multi-hop communication. The source mobile node can transmit the information to any other destination node, but it has limitations with energy consumption and battery lifetime. Since it appeals to a huge environment, there is a probability of obstacle present. Thus, the network requires finding the obstacles to evade performance degradation and also enhance the routing efficiency. To achieve this, an obstacle-aware efficient routing using a heuristic-based deep learning model is proposed in this paper. Firstly, the nodes in the MANET are employed for initiating the transmission. Further, it is needed to be predicted whether the node is malicious or not. Consequently, the prediction for link connection between the nodes is achieved by the Optimized Bi-directional Long-Short Term Memory (OBi-LSTM), where the hyperparameters are tuned by the Adaptive Horse Herd Optimization (AHHO) algorithm. Secondly, once the links are secured from the obstacle, it is undergone for routing purpose. Routing is generally used to transmit data or packets from one place to another. To attain better routing, various objective constraints like delay, distance, path availability, transmission power, and several interferences are used for deriving a multi-objective function, in which the optimal path is obtained through the AHHO algorithm. Finally, the simulation results of the proposed model ensure to yield efficient multipath routing by accurately identifying the intruder present in the network. Thus, the proposed model aims to reduce the objectives like delay, distance, and power consumption.

Fog computing is a paradigm that works in tandem with cloud computing. The emergence of fog computing has boosted cloud-based computation, especially in the case of delay-sensitive tasks, as the fog is situated closer to end devices such as sensors that generate data. While scheduling tasks, the fundamental issue is allocating resources to the fog nodes. With the ever-growing demands of the industry, there is a constant need for gateways for efficient task offloading and resource allocation, for improving the Quality of Service (QoS) parameters. This paper focuses on the smart gateways to enhance QoS and proposes a smart gateway framework for delay-sensitive and computation-intensive tasks. The proposed framework has been divided into two phases: task scheduling and task offloading. For the task scheduling phase, a dynamic priority-aware task scheduling algorithm (DP-TSA) is proposed to schedule the incoming task based on their priorities. A Memoization based Best-Fit approach (MBFA) algorithm is proposed to offload the task to the selected computational node for the task offloading phase. The proposed framework has been simulated and compared with the traditional baseline algorithms in different test case scenarios. The results show that the proposed framework not only optimized latency and throughput but also reduced energy consumption and was scalable as against the traditional algorithms.

Detecting carbon emissions is the key to carbon peaking and carbon neutrality goals. Existing research has focused on utilizing data-driven method to study carbon emissions off a single object. This study proposes a regional carbon emissions prediction method. The area objects are divided into dynamic objects for vehicles and static objects for buildings. For the dynamic object, carbon emissions is modeled using the vehicle location provided by the BeiDou satellite navigation system (BDS). For the static object, the neural network R3det (rotation remote sensing target detection) is used to identify the buildings in remote sensing images, and then the trained ARIMA time series model is used to predict the carbon emissions. The model is tested in an industrial park in Tangshan, Hebei Province in China. The result of the regional three-dimensional emission map shows that the method provided a novel and feasible idea for carbon emissions prediction.