Research On Algorithms Of Indoor Scenario Modeling Aided Collaborative Localization For Indoor Unmanned Systems

Location information is a core requirement for achieving the mobility of unmanned systems.Unmanned systems can acquire their location by integrating global navigation satellite System(GNSS)receivers.However,GNSS signals can interfere or even be completely denied in indoor environments,the unmanned systems cannot obtain more accurate indoor positions.As for indoor localization system of unmanned systems,the common integrated inertial navigation system is simple to deploy,while the localization error increases cumulatively with movement time.External wireless localization technology is one of the sources of location acquisition for unmanned systems.Although external wireless localization systems do not increase errors with movement,they are complex to deploy and easily constrained by deployment conditions and scenarios,resulting in suboptimal ranging parameters and inaccurate localization results due to the non-line of sight(NLOS)measurements.Hence,the research problems of indoor unmanned localization systems mainly include the following:(1)wireless localization system deployment,(2)wireless signal parameter measurement,and(3)location estimation and solution.To this end,this paper follows these three issues to investigate the cooperative localization method for indoor unmanned systems.In indoor scenario,access points(APs)not only take on the network access requirement,but also the task of wireless localization.The lack of deployment experts deployment experience support can lead to that is not to complete the network and localization function,but also caused a lot of waste.This paper aims at the most common Wi-Fi fingerprinting localization system,and designs an expert system for scenario modeling-based AP placement optimization.Through theoretical performance analysis,the wireless localization and network access optimization objectives and coverage constraints are proposed in conjunction with the scenario information model.The AP placement optimization is a combinatorial optimization problem,which is NP-complete,and this paper uses a heuristic algorithm and applies the Pareto front criterion and virtual repulsive force to optimize the AP deployment location.The results of the comparative experiments also demonstrate the improvement of the proposed method over the current related works and justify the improved direction of the proposed methodDifferent to fingerprint-based localization methods,wireless ranging is easily affected by obstacles in an indoor environment,especially concrete walls.Thus,the wireless ranging anchor deployment systems are also the domain of expert systems.Taking the most representative Ultra-wideband(UWB)high-precision wireless ranging and localization system as an example,this paper proposes a wireless ranging anchor placement optimization method with the collaborative participation of indoor scenario representation.The simulation technique of wireless ranging and localization simulation based on the scenario information model is constructed by improving the indoor scenario representation method.Since the anchor placement optimization problem is a combinatorial optimization problem,a heuristic algorithm is used to optimize the location of anchor placement based on the scenario information model.The results of the comparison experiments also show the improvement of the proposed method compared with the current related works,and provide anchor placement solutions with practical value for the construction personnel of localization systems.In addition,in terms of wireless signal measurement,wireless signal measurement and radio map construction is a difficult problem in fingerprint localization systems.And is also an important influencing factor to determine the accuracy and reliability of fingerprint-based localization system.To this end,we propose the wireless measurement technology of UWB/Inertial Measurement Unit(IMU)integration to construct the radio map synergistically.It replaces and reduces the error and labor cost arising from manual measurement during the wireless signal measurement and radio map construction process.The UWB/IMU co-measurement system only works in the radio map construction phase,the radio map continues to serve the AP fingerprint-based localization system after the UWB/IMU co-measurement system is withdrawn.The experiments also prove that the proposed method not only reduces the cost of radio map construction,but also guarantees the accuracy and reliability of the radio map.Finally,a wireless signal measurement method with differential ranging-based NLOS mitigation algorithm is proposed in this paper for high-precision wireless ranging-based localization systems.Since the UWB/IMU collaborative localization system is at rest and under long-term NLOS conditions,the IMU cannot participate in mitigating the NLOS ranging error,and the localization accuracy still deteriorates.The differential ranging method mitigates the NLOS error with the reference ranging.Given that indoor scenes are only used for location display and navigation path planning in many current wireless localization systems,they are not fully utilized in localization algorithm.In this paper,a cooperative indoor scenario/UWB/IMU fusion localization method is proposed to improve the localization accuracy of indoor unmanned systems under NLOS conditions.The experiments demonstrate the improvement of the proposed collaborative localization method in terms of localization accuracy and robustness.In summary,this paper investigates the wireless localization system deployment,wireless signal measurement,and collaborative localization problems with cooperative indoor scenario representation method,and proposes a new research scheme,and verifies the rationality and effectiveness of the proposed methods through experiments.