Font Size: a A A

Research On Key Technique Of Multi-Sensor Fusion Based Spherical Robot Localization In Indoor Environment

Posted on:2021-04-09Degree:MasterType:Thesis
Country:ChinaCandidate:H H HuangFull Text:PDF
GTID:2428330602486032Subject:Control Engineering
Abstract/Summary:PDF Full Text Request
The spherical robot is a new type of mobile robot.Localization is the basic problem of autonomous navigation for mobile robots.Facing the long-term stable application requirements of spherical robots in indoor environment,a multi-sensor fusion based positioning system for spherical robot platforms is studied.The main work and innovations in this dissertation are as follows:1.Explore the particularity in spherical robot positioning system design.In two different robot platforms-wheeled robots and spherical robots,the original sensor information and current positioning algorithms is compared through experiments.It is proved through experiments that the quality of the original information of the spherical robot sensor is poor,and the current positioning algorithms have poor stability and lack of accuracy in the spherical robot platform.2.Implements a wheel speed odometer algorithm suitable for spherical robots.The motion model of the platform is analyzed,the formula for recursing the robot pose is derived.The experiment proves that this method is better than the traditional wheel odometer solution,and obtains a more accurate pose tracking effect in the spherical robot.3.Implement a positioning system suitable for spherical robots.By integrating the wheel speed meter,IMU and camera,a multi-sensor fusion spherical robot positioning system is realized.Especially for the difficulty of fusion,the space-time calibration of external parameters method is designed.In engineering implementation,the correction module is realized to solve the large detection errors that may occur during the correction process.Through experiments,it is verified that the positioning system has better overall performance than pure wheel speed odometer and VINS-MONO in different environments,which verifies the applicability of the positioning system on a spherical robot platform.
Keywords/Search Tags:Spherical Robot, Vision, Indoor Positioning Algorithm, Multi-Sensor Fusion
PDF Full Text Request
Related items