A novel green software evaluation model for cloud robotics.

The energy consumption of cloud robotics is an important design factor to be considered early in system development. Software as the main enabler of cloud robots, its energy consumption will directly influence the energy consumption level of the entire system. In this paper, we propose a green software model for cloud robotics based on energy consumption time state transition matrix (ETSTM), which can effectively integrate the logic functions, energy consumption, and execution time of software into a single model. To improve the practicability of ETSTM, we provide a software energy consumption function based on software characteristic fitting by a backpropagation (BP) neural network, which can be used to predict software energy consumption. Furthermore, we provide two types of energy consumption analysis algorithms based on explicit execution path search and bounded model checking (BMC) for ETSTM. The experiment results show that the proposed method can achieve good performance. [ABSTRACT FROM AUTHOR]