Abstract
This thesis is concerned to develop design and implementation of networked control (NCS) techniques of multiple smart surveillance control systems for the internet of things. The prediction method has been popular in single agent control systems due to its capability of actively compensating for network- related constraints. This characteristic has motivated researchers to apply the prediction method to closed-loop multi-process controls over network systems. This thesis conducts an in-depth analysis of suitability of the prediction method for the control of NCS on smart surveillance control.In the multiple smart surveillance system problem, a single smart device must achieve a common output (camera position) that corresponds to the designed http protocol. The output is determined by the external reference input (different position of the camera), which is initially provided to only one smart device in the NCS. This agreement is achieved through Meta data exchanges between smart devices over network communications. In the presence of a network, the existence of network delay and data loss is inevitable. The main challenge in this thesis is thus to design an external communication protocol with an efficient capability for network constraints compensation.
The main contribution of this thesis is the enhancement of the prediction control algorithm’s capability in NCS applications in cloud control. The cloud control scheme has two control loops, one is local closed-loop control via smart device servo tracking and the other is the networked closed-loop control via cloud computing. The considered actively network communication constraints (accurate and fast tracking of moving objects) to achieve desired control performance to avoid control confliction of multiple smart surveillance systems.
In the first case, this thesis presents the designed algorithm, which is able to compensate network communication constraints for accurate tracking of moving object in linear system. The result is accompanied by stability criteria of the whole NCS an optimal gains selection and empirical data from the experimental results. In this context, current existing tracking algorithm implemented which refers to situation in the smart device experiences a delay in accessing its own information in the network. We refers a single smart device (camera) which is identical to the delay in data transfer for the neighbouring device in the same network.
In the second case, this thesis presents an extension of the designed algorithm with the enhanced capability of two smart devices control strategies. In this system, communication constraints achieved accurate and fast tracking for surveillance systems control. We implement machine learning DNN module training feature for accurate and fast tracking which is improve significant efficient for the data optimization feedback control system.
In the third case, we present a further enhancement of the designed control algorithm, which includes the multiple smart devices control system. It has significant importance in practical application and cloud control strategies. The communication constraints achieved by optimal feedback control by machine learning DNN module.
In the final case, we propose a novel strategy for combining the predictive control algorithm for the cloud control system, which is local closed loop control, and coordinate cloud control via optimal coordinate control (broadcast domain).
In each case, the designed control algorithm is compared with proportional Integral (PI) controller. We conduct an evaluation of the smart surveillance system output performance for each by analytical calculations, and practical experiments. We accomplished the research work in this thesis through the integration.
| Date of Award | 2024 |
|---|---|
| Original language | English |
| Supervisor | Dr Hammad Nazir (Supervisor), Iain Shewring (Supervisor) & Ali Roula (Supervisor) |