Exploration And Map Building Of Unknown Region For Mobile Robot Based On ROS

It will form high temperature,smoke,gas,collapse and other dangerous environment so that the rescue workers are too difficult to enter in the rescue missions of fire,mining accident,earthquakes and other disasters.Mobile robot can replace the rescue workers complete the task and reduce casualties,it has received wide attention,what is more,it has positive scientific value and practical significance.This topic will be based on the background of using the mobile robot to explore the unknown environment.Using the opening source Robot Operating System based on Linux,the two wheeled mobile robot enters the indoor disaster scene of unknown environment from any initial position,uses the airborne lidar to explore the environment and obtains knowledge about the environment,what is more,it explains the scene,builds appropriate maps and establishes global coordinate system.In addition,it finds its positioning relating to those maps,identifies the accessible area to choose the best path to complete the whole maps and deploys the wireless sensor's network node.It also can transmit the environmental model which has been established to a remote PC or a handed terminal in real time.In order to strengthen the movement control ability of the mobile robot,this topic uses a design called ROS robot operating system and constructed a mobile robot platform.For the hardware,it adopts Irobot differential drive chassis and UTM30LX laser radar.Through the establishment of mathematical model and analysis,we know the calculation method only based on the odometer's track has a large error and it will increase monotonically over time.Consequently,this topic corrects it according to the environmental feature and makes the system tend to be stable by the continuous iteration of the extended Calman filter.What is more,the error covariance does not change with time.On this basis,it proposes the synchronization localization based on the Hector SLAM and the composition algorithm for the indoor environment.By using the data matching of laser radar,it can have a positioning successfully and have a connection from local map to global map.The experimental results show that the constructed mobile robot platform has good robustness and can construct two-dimensional plan of the indoor environmental accurately.What is more,it can achieve higher accuracy for the presence of the interference of the ground.For the path planning,it puts forward an improved path planning algorithm of genetic based on the A*algorithm.It makes full use of A*algorithm and searches for a shortest path.In addition,it also can use the genetic algorithm to optimize the initial path.Combining the two algorithms,we can get the optimal or suboptimal path of the robot mobile.As for the problem of the limited distance in the traditional remote control and the wireless control,this topic designs and achieves the wireless sensor network's multihop communication link.By using the function of the WSN's ad-hoc network,it can form a robot's control link which is greater than the traditional way.This robot's control link can meet the needs of the disaster relief in the complex environment as much as possible.At the same time,the wireless sensor network node can carry temperature,humidity,gas concentration and other sensors to feel the information of the around environment in real-time.Combining the Robot's maneuverability and strong computing power and the wireless sensor network's communication distance,distributed computing,environmental awareness and so on,it can improve the efficiency of the rescue mission significantly.