Robot vision SLAM closed-loop detection method based on stack type combined auto-encoder

Abstract

The invention discloses a robot vision SLAM closed-loop detection method based on a stack type combined auto-encoder, and belongs to the field of mobile robot vision SLAM. The method comprises the following steps: S1, preprocessing a visual SLAM scene image, and inputting the visual SLAM scene image into a stack type auto-encoder model; s2, training a network model layer by layer, iterating network parameters by adopting a stochastic gradient descent algorithm, and continuously adjusting the model parameters to minimize a reconstruction error; s3, extracting a feature vector of the visual SLAMscene image by using the trained stack type combined auto-encoder; s4, calculating the similarity between the feature vector VK of the kth key frame (current frame) of the visual SLAM scene and the feature vectors V1, V2,..., VN of the historical key frames; and S5, comparing the similarity score with a set threshold value, and if the similarity score is greater than the set threshold value, judging that a closed loop is formed. According to the invention, the accuracy and robustness of visual SLAM closed-loop detection can be effectively improved.

Description

technical field [0001] The invention belongs to the field of mobile robot visual SLAM, in particular to a visual SLAM closed-loop detection method based on a stacked combined autoencoder. Background technique [0002] Simultaneous Localization and Mapping (SLAM) refers to the real-time positioning and construction of quantitative environmental maps by robots during their movement in an unknown environment. Visual SLAM uses a camera as a sensor to build a three-dimensional environment map in real time. A complete visual SLAM system mainly includes four modules: front-end visual odometry, back-end nonlinear optimization, loop closure detection and mapping. Loop closure detection is a key module in visual SLAM, which plays a very important role in eliminating accumulated errors. Closed-loop detection refers to judging whether the robot has returned to a certain position that already exists in the map when the current observation information and map information are given. Mos...

AI Technical Summary

Problems solved by technology

Traditional stacked autoencoders are usually stacked with multiple layers of the same autoencoder. This kind of network is easy to lose features or cause dimensionality explosion problems.

Images