| When firefighters perform operations and rescue in an unfamiliar interior space,they cannot know exactly where they are in fact.It is difficult to carry out rescue work once they encounter danger.Therefore,obtaining firefighters’ exact position in real time is an important measure to ensure their own safety.Existing positioning technologies such as GPS/ Beidou,Wi Fi,UWB,Zig Bee,RFID,etc.,are unable to meet the requirements of firefighters’ accurate localization in the unknown and complex interior space,due to their respective defects and shortcomings.Owing to its features such as no need to deploy infrastructure in advance in interior space,fast and accurate localization and so on,more and more attention in the research of indoor positioning technology has been paid to inertial navigation technology recently.Aiming at the positioning problem of firefighters in unfamiliar and complex interior space,a three-dimensional localization system for interior fireground based on inertial navigation is studied and devised in the thesis.First of all,the localization requirements of firefighter indoor disaster relief process are analyzed,and a portable positioning device for individual firefighter wearing on the waist is designed,that contains power supply system,microprocessor,inertial measurement unit,wireless transmission unit,etc.,which is able to accomplish the collection,calculation and transfers of messages such as step frequency,step length and heading angle and so on.Then,the PDR algorithm of pedestrian dead reckonings was carried out in-depth research.In view of the complex motion conditions such as going up and down stairs and slopes in step frequency detection,a peak-to-peak value detection method is used to replace the peak detection method,which can availably lessen the probability of error decision.When calculating step size estimation,different step size estimation coefficients are used to adapt different speed,which reduces the error of step size estimation.Secondly,in the judgment of the firefighters’ heading angle,the performance of Heuristic Drift Elimination(HDE)algorithm can no longer meet the needs of indoor positioning and navigation in modern buildings,this thesis presents an improved Heuristic Drift Elimination algorithm(IHDE),which uses different feedback coefficients to dynamically adjust and correct heading angle according to the type of walking path.Then,in the calculation of personnel position,for the sake of determining the location of firefighter synchronously,a dynamic location reckoning algorithm based on three-window sliding window is designed.In addition,Lora radio frequency communication technology that can transfer the location data from the personal equipment to the monitoring center,and a semi-duplex polling communication mechanism is devised to complete discovery of the multiple firefighters’ location in real time simultaneously.Finally,the software to map the firefighter’s trajectory is compiled with Python and other language,which can describe the current position coordinates of the firefighters,and save,replay the three-dimensional movement track.To sum up,for multi-storey,up and down stairs,turning,straight and other complex walking scenes with a variety of turning angles,the system can achieve exact localization of several simulated firefighters.Statistical results of many test data indicate that the designed algorithms have an excellent adaptability and robustness to various complex interior space,and the relative horizontal localization error obtained by the system is not greater than 0.5%,which can better meet the needs of interior fireground positioning. |
