Deep neural network has been widely used in image analysis, speech recognition, target detection, semantic segmentation, face recognition, automatic driving, and other fields due to its excellent algorithm performance. In this research, the neural network is mainly used to simulate robot target localization and visual navigation. Firstly, the data set based on ICP algorithm is constructed, and then, the navigation and positioning of robot are simulated by using neural network. The results show that the combination of ICP algorithm and direct method can effectively solve the problem of pose loss and expand the indoor range of data collection. The method using artificial neural network is effective and has better robustness and stability than the method based on frame matching. From the perspective of vision, positioning and navigation based on a single RGB image are feasible, and the processing time is relatively short, which can meet the requirements of real time and has higher practical value in general indoor scenes.

In order to improve the accuracy of robot control system, a scheme based on artificial intelligence is proposed. On the basis of the software environment of reinforcement learning simulation platform, a kind of rounding scheme in dynamic environment is designed and simulated. The results show that when the inclination sensor is placed on an inclined plane of 300 and collected for ten times, the maximum error of measurement that can be seen from the experimental data is 0.40. The relative included angles were 30°, 45°, 60°, and 90°, respectively, by compass sensor. The measurement was carried out, and the average value of each angle was measured 5 times. It can be seen from the experimental data that the measurement error meets the requirements of the system. Therefore, it is feasible to use artificial intelligence algorithm to optimize the robot control system.