Robot master control system

A robot master control system, comprising:

• a master controller, configured to control at least one dual-robot control system, wherein each of the least one dual-robot control system includes a first robot, a second robot, and a sub-controller controlling the first robot and the second robot, and
• the sub-controller is controlled by the master controller, and configured to interact with a preset neural network to obtain boundary information of objects in a first image, wherein the first image is transmitted by an image capturing device of each of the at least one dual-robot control system, the neural network includes a hierarchical convolutional network,
• the master controller includes:
• at least one storage medium storing at least one set of instructions, and
• at least one processor in communication with the at least one storage medium, wherein during operation, the at least one processor executes the at least one set of instructions to control the at least one dual-robot control system, the hierarchical convolutional network includes:
• a first branch performing a first feature extraction on an inputted image to obtain a first feature extraction result,
• a second branch receiving the first feature extraction result after a feature filtering operation, and performing a second feature extraction to obtain a second feature extraction result,
• a fourth branch connected to the first branch, and
• a fifth branch connected to the second branch, wherein the at least one processor executes the at least one set of instructions to:
• divide the fourth branch into regions,
• for each region, determine an importance degree based on the first feature extraction result and corresponds to each region to obtain a first importance degree corresponding to the region,
• erase, from the first feature extraction result, information corresponding to a region with a first importance degree greater than a first preset threshold so as to obtain a first correction information,
• output, by the fourth branch, the first correction information, and
• input the first correction information to the second branch.