Research of Mobile Robot on Indoor Environment Exploration Technology Based on Laser Radar
|School||Central South University|
|Course||Pattern Recognition and Intelligent Systems|
|Keywords||mobile robot autonomous exploration laser radar feature extraction map building|
With the continued research and application of robot technology extensively, an autonomous mobile robot is demanded increasingly to work under unknown environment. Autonomous exploration is a very important part of mobile robot technology, and is the ability that robot must have to successfully work under unknown environment. It bases on reliable mobile robot experiment platform, and efficiently explores by control system under complex environment. It is a hot topic concerning to the current international academic community.This paper has mainly studied on the related technology based on laser radar of autonomous exploration of mobile robot, it taked the mobile robot MORCS-1 as experiment platform, which is developed by Central South University Intelligent Lab. The main contribution and work of this paper are described as follows:1 Designed a set of algorithms of feature extraction for these typical environmental characteristics such as wall, corner, passageway, room and corridor etc. The algorithm takes full advantage of fast and accurate characteristics of laser radar, mapped the sensor data in the robot local geometric map to analyze. Ranging data from the laser radar in a sampling time is used to obtain the environment characteristics, operation is simple and fast. It can scan and analyze the environment many times in a short time, meet the demand for robot real-time navigation, has the improvement and the enhancement compared with the traditional algorithm performance.2 Established a mixed environment model which combine geometry model with topology. Built geometry map by environmental features which was extracted from data of laser radar. And then extracted topology information based on the geometry map, choosing the environmental characteristic point as topology node to build out the topology map, every topology node preserves its detailed geometry information of corresponding area at the same time, then the mixed model perfectly describing the environment from geometry and topological way.3 Aimed at exploring the interior unknown environment, designed a high-effect strategy which combined the geometry with topology: Firstly, robot observed environment at the current node to extract candidate nodes, then evaluate these nodes synthetically from both geometry and topology, choose the optimal node as the next step will be visited, added those nodes which have been visited to topology model at the same time. Specifically designed corresponding robot behaviors for the different topology node, to make a robot can come to choose corresponding behavior according to different node. Finally, designed a Finite State Automata to control and plan the specific process of robot’s behaviors conversion at different nodes, to enable the entire process of exploration to be smooth and orderly.The validity of this technology was validated by a lot of experiments on the mobile robot MORCS-1, providing certain experience to go deep into studying various technologies of robot exploration.