Simultaneous Localization And Mapping Technique Combied With Scene Detection Based On Multisensor

As the industrial robots and service robots are advancing by leaps and bounds,top priority has been given to robotics due to its significance in development of society and people's living standard.Simultaneous localization and mapping(SLAM)has been widely regarded as a key part of automatic robot because it gives a main solution to where robots are and what the environment is like.Different sensors have differnet capacities and may even fail in different scenes.This will influence reliability and accuracy of SLAM systems which utilize those sensors.To enhance the robustness of SLAM systems,we present multi-scene SLAM technique in this paper,which is a novel extensible SLAM framework by combing different SLAM systems facilitated by a scene detection method.It can activate different SLAM model automatically based on the scene detection result.The main contributions of our paper are in four fold:Firstly,we propose a robust and extensible SLAM framework called SceneSLAM to enhance the self-adaptive performance of current SLAM systems.By model structure design,it is easy to utilize open-source SLAM systems to be SLAM models of SceneSLAM and modify the scene detection models to support complex application scenarios.Based on the scene detection results,it can schedule the SLAM models automatically.Secondly,aiming at indoor,outdoor and dark scene detection problems,we present a scene detection model based on convolutional neural network and Bias filter optimization.Experimental results show that the proposed scene detection model has a certain accuracy and stability in detecting indoor,outdoor and dark scenes.What's more,considering sensor failure in indoor,outdoor and dark scenes,we design Laser SLAM model and Vision SLAM model to handle the issues.The Laser SLAM model is used to a dark scene for mapping and localization tasks.The Vision model includes monocular pattern and depth pattern which are applied to deal with outdoor scene and indoor scene respectively.Map fusion is performed to achieve globally consistent localization and mapping when scheduling SLAM models.At last but least,on the basis of study above we build a prototype system based on SceneSLAM to enhance the reliability of existing SLAM systems when the scene changes between indoor,outdoor and dark scene.The prototype system runs on a TurtleBot robot equipped with a Kinect sensor.The experimental results show that multi-scene SLAM technique proposed in this paper provides an effective solution to enhance the robustness of existing SLAM systems in dynamic environments.