Robust slam method for indoor robots combined with environmental semantics

Abstract

The present invention provides a robust slam method for indoor robots combined with environmental semantics. The method comprises: indoor robot positioning of the fusion of GPS global signals, IMU inertial measurement unit signals and lidar signals; an indoor robot perception of the fusion of radar data image data, radar sensor calibrated external parameters, pre-camera calibrated external and internal parameters, and a host speed and angular velocity; an indoor robot semantic map of the fusion of the NavigationInfo of a dreamview module, a LaneMarker from a sensing module, and positioning information of a positioning module; decision making of the indoor robot of the fusion of obstacle information, vehicle status, traffic lights, and map information; and indoor robot path planning for thefusion of positioning, a perception, prediction, routing, and a map. According to the method provided by the present invention, the robot can be more robust to the environment in motion and decisionmaking in the environment.

Description

technical field [0001] The invention belongs to the technical field of computer vision, and in particular relates to a robust slam method for indoor robots combined with environmental semantics. Background technique [0002] With the rapid development of artificial intelligence technology, especially strong intelligence, robot application scenarios have gradually shifted from extreme environments and industrial application scenarios to indoor work scenarios that closely interact with humans. Based on this, our design and use of robots are gradually shifting from functionality to robustness, but existing indoor robots can meet the needs of indoor high-precision positioning and navigation in a simple environment, but once the environment changes or there is high noise, Robot vision will produce cumulative errors at different scales, which is very unfavorable for robot navigation; in addition, indoor robots that have entered human society cannot understand environmental informa...

AI Technical Summary

Problems solved by technology

Based on this, our design and use of robots are gradually shifting from functionality to robustness, but existing indoor robots can meet the needs of indoor high-precision positioning and navigation in a simple environment, but once the environment changes or there is high noise, Robot vision will produce cumulative errors at different scales, which is very unfavorable for robot navigation; in addition, indoor robots that have entered human society cannot understand environmental information, so completing specified tasks requires the support of follow-up path planning. It cannot be called truly "strong" intelligence.

[0003] Logistics robots can complete intelligent warehousing and automated logistics indoors. However, due to the fact that the sealed warehousing environment is often dominated by white backgrounds and lacks texture features, this will cause scale information loss during robot navigation and damage the positioning accuracy of the visual odometry. ; and the present invention integrates the environmental semantics into the map and uses optimization means such as potential field guidance to obtain a robust slam method that can read the environmental map

Images