This paper explores the utilization of a novel transformer-based architecture for end-to-end learning in predicting steering angles in self-driving scenarios while leveraging a novel robust image processing pipeline t...
This paper explores the utilization of a novel transformer-based architecture for end-to-end learning in predicting steering angles in self-driving scenarios while leveraging a novel robust image processing pipeline to handle diverse environmental situations. Our approach relies solely on visual perception as the input to generate control commands. We trained and evaluated our methodology using a proprietary dataset from a self-driving car simulator consisting of image frames paired with their corresponding steering angles. The presented methodology is robust against overfitting, and it shows superior performance in terms of Mean Squared Error (MSE) and Mean Absolute Error (MAE) compared to previous methods.
Decentralized Federated Learning (DFL) enables collaborative model training without a central server but faces challenges in efficiency, stability, and trustworthiness due to communication and computational limitation...
详细信息
Human Activity Recognition (HAR) is essential in various applications, including wellness tracking, automated residences, and fitness monitoring. In the past few decades, sensor-based HAR has become increasingly popul...
详细信息
The development of a distributed trajectory-tracking control strategy that is independent of velocity measurements is critical to achieving finite-time tracking control of autonomous underwater vehicle(AUV) systems. I...
详细信息
The development of a distributed trajectory-tracking control strategy that is independent of velocity measurements is critical to achieving finite-time tracking control of autonomous underwater vehicle(AUV) systems. In this study, a group of heterogeneous AUV systems with intermittent communication links is considered and a finite-time trajectory-tracking control strategy is developed. The strategy includes two observers and one controller proposed for each follower-AUV. The first observer, a hybrid finite-time observer,estimates the state of the leader, whereas the second observer, which relies only on the position measurement,is proposed to estimate the states of the follower-AUV itself. In addition, a distributed trajectory-tracking controller is designed using the states estimated by the intermittent communication network even without velocity measurements. A homogeneous technique is utilized to prove that all followers can track the leader in a finite time. Finally, the effectiveness of the developed finite-time tracking control strategy is illustrated by numerical simulations.
Resource optimisation is commonly used in workload management, ensuring efficient and timely task completion utilising available resources. It serves to minimise costs, prompting the development of numerous algorithms...
In [3] it is shown, answering a question of Jordán and Nguyen [9], that universal rigidity of a generic bar-joint framework in R1 depends on more than the ordering of the vertices. The graph G that was used in th...
详细信息
Dear editor,We consider a containment control problem for a linear discrete-time (DT) multi-agent system via reinforcement learning. Containment control(CC) of multi-agent networks has received extensive attention in ...
详细信息
Dear editor,We consider a containment control problem for a linear discrete-time (DT) multi-agent system via reinforcement learning. Containment control(CC) of multi-agent networks has received extensive attention in the control community in recent years [1–4]. CC is motivated by natural phenomena and has potential and vital applications in practical engineering. For example, one applica-
Uniform expressivity guarantees that a Graph Neural Network (GNN) can express a query without the parameters depending on the size of the input graphs. This property is desirable in applications in order to have a num...
详细信息
The dominant intention of this article is to extract the new exact traveling waves solutions of the nonlinear space-time fractional Sharma-Tasso-Olver equation in the sense of beta-derivative by using three integratio...
详细信息
Heat transport analysis for non-Newtonian fluid flows between non-parallel wall channels has sustainable significance in high-performance thermal engineering processes. In recent years, this analysis is extensively us...
Heat transport analysis for non-Newtonian fluid flows between non-parallel wall channels has sustainable significance in high-performance thermal engineering processes. In recent years, this analysis is extensively used in numerous natural flows and industrial processes, for instance, blood flow through human veins, lubrication systems, automobile radiators, thermal pumps, and water purification processes, etc. Therefore, this research, it is targeted to enhance thermal performance with the addition of ultrafine metallic nanoparticles into working fluids. With this goal in mind, this research work presents a numerical investigation for buoyancy-driven flow of Carreau nanofluids confined in a vertical converging enclosure. In addition, heat and mass transport analysis with non-linear thermal radiation and activation energy are mathematically formulated via Buongiorno’s model. A new formulation is developed for purely radial flow inside this converging channel and appropriate non-dimensional variables are utilized for problem simplification. These transformed equations are then numerically tackled with the help of a versatile numerical method, bvp4c function in MATLAB. The simulated results are portrayed by virtue of nanofluid velocity, temperature, and concentration distributions with variation in governing dimensionless parameters. The results indicate that the velocity was significantly reduced with higher activation energy parameter. Moreover, the higher values of the Grashof number yields increasing conduct in velocity distributions.
暂无评论