Method and device for segmenting drivable road surface in an indoor environment, medium and robot

Abstract

The invention discloses a method and device for segmenting a drivable road surface in an indoor environment, a medium and a robot, and the method comprises the following steps: collecting indoor road surface pictures, and preprocessing the road surface pictures to form a training set; constructing a pavement segmentation network model; training the pavement segmentation network model by using the pictures of the training set; and collecting an indoor real-time road surface picture, and detecting the road surface position and the road surface size in the real-time road surface picture based on the trained road surface segmentation network model. An existing semantic segmentation algorithm and an existing semantic segmentation model are improved in a targeted mode, meanwhile, a Softharding strategy is adopted, the network over-fitting phenomenon is relieved, the network structure is simple and efficient, the calculation speed is high, and therefore the accuracy and real-time performance of road surface segmentation are guaranteed, the robot is better assisted in walking safely indoors, and the robot can walk more safely. And the danger that the robot falls from steps, escalators and other areas is reduced.

Description

technical field [0001] The invention relates to the field of robots, in particular to a method, device, medium and robot for segmenting a drivable road surface in an indoor environment. Background technique [0002] With the development of robots, indoor mobile robots have begun to shuttle in indoor areas such as office buildings, hotels, and shopping malls. How to ensure that robots can safely walk in indoor areas has always been a difficult problem in the industry. During the walking process of the robot, sensors such as lasers and ultrasonic waves can be used to detect surrounding obstacles, such as walls, gates, doors, etc. However, steps and escalators in office buildings or shopping malls can easily cause the sensor of the robot to fail, causing the robot to fall from a height. [0003] Semantic Segmentation (Semantic Segmentation) technology is a very important topic in today's computer vision. The main function of this technology is to classify different categories ...

AI Technical Summary

Problems solved by technology

The existing semantic segmentation methods usually have a very large amount of calculation, and the computing power of edge devices such as robots is limited, and it usually takes more than 1 second to process a picture. Such a high delay makes it difficult for the existing semantic segmentation technology to be used in practice. use in the scene

Images