Research And Implementation Of Swarm Localization Using Ultra-wideband

With the development of robots and mobile devices,robots and devices are becoming smaller and smarter.As human-machine collaboration and machine collaboration emerging,more applications will use swarms of multiple robots and devices to work collaboratively in the future.Therefore,localization is important for nodes in the swarm.Ultra-wideband(UWB)technology is one of the most popular technology for indoor localization,which has high precision and good robustness.A device or a robot usually needs to exchange information with multiple anchors when localization.However,a node sometimes encounter locating failure due to the lack of nearby anchors.As a result,in an environment where anchors are sparse,how to perform localization for a node within a swarm still remains a problem.This thesis aims to address the problem by using the inter-distance among nodes within a swarm for sparse nearby anchors.However,We still face the following challenges: on the one hand,exsting ranging schema is lack of efficiency and not suitable for swarm;on the other hand,how to achieve position awareness with only these information.To address the above problem,this thesis conducts the contribution as follows:(1)This thesis designs a new UWB ranging schema called UWB swarm ranging to improve the ranging efficiency in a swarm.The message used in swarm ranging is called ranging message which is broadcasted periodically.Since the message is broadcasted,the efficiency is improved.Experiments show that the swarm ranging has high ranging efficiency,good robustness,and good adaptability.(2)This thesis proposes a distance-based algorithm for position awareness based on swam ranging and extended Kalman filter.The position awareness problem is modeled and analysed properly while a implicit function between yaw and distance is found.With the help of the position awareness algorithm,nodes can fuse the distance information and other sensor information to get a good state estimation.Experiments show that nodes can sense their position in real-time.(3)A position awareness system is designed based on these technologies.The experiment shows the system performs well.In short,this thesis solves the position awareness problem by proposing a swarm ranging schema and position awareness algorithm.The research results may address the position awareness problem in a sparse anchor environment and provide a fundamental localization schema for swarm-based applications in the future,which is of great social and economic significance.