| With the rapid development of information and the acceleration of life and work,people need to get rid of complicated family work.At the same time,after the implementation of the multi-year family planning policy in China,the resulting population-inverted pyramid structure has made the aging family structure more and more,which will inevitably increase the pressure on more young families.The pace of life and work pressure will also accompany young people.Children’s time is getting less and less.The service robot market will follow and will be widely favored by people.Intelligent positioning in complex environments is a prerequisite for service robots to provide high quality services to humans.The positioning of service robots is more indoor positioning.Due to the complexity of the indoor environment and the accuracy of positioning and special requirements for safety,indoor wireless positioning technology has different characteristics from ordinary positioning systems.These features are used in outdoor positioning technology.It is not available.How to choose reasonable and effective positioning methods in various indoor environments has become the key to research service technology robot positioning technology.However,since most models and systems are developed based on a single data source,positioning accuracy is low;some high-precision algorithm models based on a single data source require expensive hardware facilities to support.Therefore,the use of multi-source information fusion method to achieve the complementary advantages of various positioning technologies and reduce hardware costs has become the focus of indoor positioning research.In view of the problems existing in the existing wireless positioning technology,this paper selects the Wi-Fi wireless communication technology with high security,low power consumption,low cost and many hot spots as the research object.The main work and innovations are as follows:First,when wireless signals are transmitted in a complex and variable environment,the effects caused by various factors,especially Wi-Fi hardware devices,cannot be avoided,resulting in insufficient coverage of the wireless signal and attenuation of the signal strength.In this paper,an ultra-small Wi-Fi antenna unit is proposed,which uses DGS structure as the reflector of the antenna,and integrates the antenna working at 1.7 GHz ~ 2.7 GHz with the DGS structure of the corresponding working frequency.The designed Wi-Fi antenna effectively provides a compact structure,a wide frequency,a variety of layouts and omnidirectional coverage,thus more effectively ensuring that the signal strength attenuation of the wireless signal is finally reduced,and the propagation efficiency is guaranteed.When the positioning accuracy is improved,the positioning cost can be reduced,and the high-precision positioning requirement of the service robot can be achieved.Secondly,it is difficult to construct the fingerprint feature database in the positioning algorithm,and the matching data volume is large,the calculation time is long,and the real-time performance is not good.In this paper,a fast location algorithm based on location fingerprint is proposed.The cluster fingerprint database is established in the data collection stage,and then the cluster subset is selected.Then the fingerprint points are selected from the subset and the position distance is estimated to reduce the computation time and fast.Estimate the purpose of the position distance.In the position estimation ranging,the data with large variation range is standardized and weighted,and overcoming the same weight will seriously lead to the sample imbalance problem.Improve the fingerprint matching algorithm without changing the positioning accuracy or improve the positioning accuracy,improve the accuracy of the data in the fingerprint database,reduce the complexity of the fingerprint database,effectively reduce the positioning accuracy to the fingerprint feature point density,and improve the positioning.The positioning of the system is real-time,reducing the power consumption of the service robot.Thirdly,for the wireless signal propagation,the noise interference factors are mostly,the single information source positioning is seriously interfered,and the positioning accuracy is not high.In this paper,a multi-source information fusion localization algorithm based on extended Dempster-Shafer evidence reasoning is proposed.Dempster-Shafer evidence reasoning is used to solve the indoor location problem based on RSSI location fingerprinting method.The basic probability is constructed by establishing an identification framework that meets the positioning requirements.The allocation function,then according to a certain combination of rules,the RSSI values from different information sources are merged to establish a decision rule to obtain the final target position.Solve the obvious shortcomings of the traditional Dempster-Shafer evidence reasoning in solving the serious conflict of evidence.That is,the problem one is that when the conflict between the evidences is low and the set bases are very different,the distribution of trust will be inappropriate.Question 2: When the conflict between the evidences is high and cannot be effectively dealt with,the conclusions often obtained are inconsistent with common sense.By integrating information from different information sources,comprehensive information that cannot be obtained by a single information source can be obtained,thereby effectively improving positioning accuracy.Fourthly,in view of the complex shape of obstacles in the actual environment and the poor obstacle avoidance results of serviced robots after precise positioning,this paper presents a global optimal path planning obstacle avoidance method for convex polygon irregular obstacles.After the application of the algorithm is extended,the environment perception model is first established.Then the idea of heuristic algorithm is used to propose an improved geometric path planning obstacle avoidance algorithm.The performance of the algorithm is analyzed by simulation and simulation.After the positioning,when the service robot encounters complex obstacles,the algorithm can satisfy the global optimal path planning of the convex polygon obstacles,thereby improving the accuracy of the search path. |
PDF Full Text Request