In the current operating modes used in working of live robots in industries, we cannot fulfill the computational power requirements needed by AI highly-precise algorithms. It's also challenging to handle the complex and large working environment, and cannot be used in a situation where a N number of robots are working simultaneously and generating huge data which has to be managed and controlled. In light of these flaws, this study compares and analyses both the operating modes of robots which are used in power grid inspection and working of robots used in industries, and presents an intelligent control technique for managing live working robots using cloud and edge computing module. The challenges that existed in the previous can be solved by deploying distinct components of the deep learning algorithm in cloud computing and edge terminals, the issues with working mode have been resolved. An example of live working robot situation in the substation is investigated. The findings reveal that the proposed intelligent control approach in this paper can meet the requirements. Artificial intelligence(AI), deep learning requires a lot of learning, computing power, and Self-learning, recognition, self-renewal, coordinated neural network control and other highly-precise control algorithm, and it can be used in a scenario with a lot of moving parts or a vast robots working simultaneously which generates large amount of data to be managed and controlled.