Deep cross altitude visual interpretation for service robotic agents in smart city

Multi-agent robotic platforms are increasingly used for various commercial applications. In this paper, a cross altitude visual analytic framework for a group of robots, singularly referred to as MOdular RApidly Deployable Decision Support Agent (MORAD DSA), used for decision support and various services in the smart city environment is presented. The robotic subsystem consists of two agents operating in different altitudes. These agents give the decision support system the ability to have encompassing view of the operating environment. The visual analytic system which is the focus of this paper uses a deep convolutional neural network to learn the complex patterns required by the urban management responsibilities. Several smart city applications such as sidewalk pavement inspection, sidewalk sweeping, rubbish detection and parks management scenarios are used for real world simulation of the proposed framework. The experimental results show that the proposed algorithm is capable of performing complex inspection and decision-making tasks required for smart city management.