Search

This paper proposes a resource allocation method based on deep learning in power internet of things. The method uses density-based spatial clustering of noisy applications (DBSCAN) to analyze data, and then trains deep neural networks (DNN) based on the results of DBSCAN data. The results show that our method can significantly improve the user's quality of service (QoS).

This paper presents the dynamics and motion control of a two-wheeled mobile robot (2WMR). The 2WMR is driven by two independent wheel motors, upon which an intermediate body (IB) is mounted. Based on Lagrange Equations, the mathematical model of the 2WMR is derived. The 2WMR having two actuated configuration variables (left wheel angle and right wheel angle) and one unactuated configuration variable (tilt angle of the IB) is a typical underactuated system. And it is inherently unstable and all the configuration variables have to been controlled through the actions of the wheel motors. Based on linearized model of the deduced dynamics, a double-loop proportional-derivative (PD) control strategy is designed to balancing control of the IB and follow the desired wheel speed motion. Moreover, a more robust double-loop fuzzy control scheme is proposed according to the parameters uncertainty of the 2WMR. The universe values of the fuzzy controller are selected based on the simulation of the double-loop PD controller. Simulation and experimental results are presented to demonstrate the feasibility and effectiveness of the proposed motion control design for the 2WMR.