To solve the problems of poor exploration ability and convergence speed of traditional deep reinforcement learning in the navigation task of the patrol robot under indoor specified routes, an improved deep reinforcement learning algorithm based on Pan/Tilt/Zoom(PTZ) image information was proposed in this paper. The obtained symmetric image information and target position information are taken as the input of the network, the speed of the robot is taken as the output of the next action, and the circular route with boundary is taken as the test. The improved reward and punishment function is designed to improve the convergence speed of the algorithm and optimize the path so that the robot can plan a safer path while avoiding obstacles first. Compared with Deep Q Network(DQN) algorithm, the convergence speed after improvement is shortened by about 40%, and the loss function is more stable.

In this paper, a voltage-boost-type non-voltage drop single-phase full-bridge inverter connected to a switched-capacitor structure is proposed. The output voltage of the inverter is controlled by the pulse width modulation of a DSP to control the lead and break of the active switches. The full-bridge switches work at low frequency; the other switches work at high frequency. The inverter uses two capacitor modules to charge and discharge alternately so as to overcome the problem of voltage drop on the output side of the inverter in the transition stage from series capacitor discharge to parallel charge. By analyzing the charge–discharge characteristics of the RC charge–discharge circuit, the capacitor charge–discharge cycle can be adjusted to alter the output voltage within a certain range. The results from the physical construction verify the Simulation results achieved well, which demonstrates satisfactory performance that supports the verification of the above theory.