SBIR Phase I: Mobile Indoor Localization and Navigation System Using Sensory Data with Data Mining and Machine Learning Techniques

这个小型企业创新研究（SBIR）第一阶段项目将探索一种室内定位技术的可行性，该技术在GPS无法工作的情况下可以高效准确地工作。GPS接收机虽然被一般公众普遍采用，但不能在室内工作，并且在室外城市环境中存在高达25米的误差。 通过分析和处理来自WiFi、蓝牙、手机信号、磁力计、加速度计、指南针和陀螺仪的数据，可以创建建筑物的感官蓝图。然后，可以利用建筑物的感官蓝图来定位持有智能移动的设备的人。研究将包括调查，设计，实施和验证以下内容：（i）能够用加速度计、罗盘和陀螺仪检测用户的运动的运动模型，（ii）要在速度和定位精度方面进行比较的不同机器学习算法的不同集合，（iii）从阶段（ii）的最佳机器学习算法构建的概率测量模型，以及（iv）组合运动和测量模型的蒙特卡罗定位（MCL）算法。最终的室内定位算法将被实现，在真实世界的条件下进行测试，并进行完善，以证明该技术在准确性，速度和适用性方面的优越性。该项目更广泛的影响/商业潜力在于它可以彻底改变建筑物的使用方式。该技术为最终用户和公司都带来了好处。一方面，大型建筑物内的用户（例如，超级市场、购物中心、医院、博物馆）将能够直接在其移动的设备上访问楼层平面图、基于位置的信息和逐向导航。另一方面，公司将能够分析客户的动向，并在他们需要的时候和地点为他们提供有针对性的信息或广告。该技术的其他应用将带来社会效益：（i）第一反应者将能够准确地确定受害者的位置，从而缩短反应时间并拯救生命，（ii）建筑物管理人员将能够通过根据房间的占用情况实时调节每个房间来节省高达30%的能源和资金，（iii）残疾人将能够使用该技术来提供帮助，例如寻找轮椅无障碍路线，及（iv）仓库经理将可缩短履行订单的时间。在不久的将来，室内定位将像今天的GPS一样普及。

This Small Business Innovation Research (SBIR) Phase I project will explore the feasibility of an indoor localization technology that works efficiently and accurately when GPS does not. GPS receivers, although universally adopted by the general public, do not work indoors and suffer from inaccuracies of up to 25 meters in outdoor urban environments. By analyzing and processing data generated from WiFi, bluetooth, cellphone signals, magnetometers, accelerometers, compasses, and gyroscopes, a building's sensory blueprint can be created. The building's sensory blueprint can then be exploited to localize people holding smart mobile devices. The research will consist in investigating, designing, implementing, and validating the following: (i) a motion model capable of detecting a user's movement with accelerometers, compasses, and gyroscopes, (ii) a diverse set of different machine learning algorithms to be compared in terms of speed and localization accuracy, (iii) a probabilistic measurement model built from the best machine learning algorithm of phase (ii), and (iv) a Monte Carlo Localization (MCL) algorithm that combines the motion and measurement models. The final indoor localization algorithm will be implemented, tested in real-world conditions, and refined to prove the technology's superiority in terms of accuracy, speed, and applicability.The broader impact/commercial potential of this project is that it could revolutionize the way buildings are used. The technology offers benefits to both end-users and companies. On one hand, users inside large buildings (e.g., supermarkets, shopping malls, hospitals, museums) will have access to floor plans, location-based information, and turn-by-turn directions directly on their mobile devices. On the other hand, companies will be able to analyze their customers' movements and provide them with targeted information or advertising when and where they need it. Other applications of the technology will provide societal benefits: (i) first responders will be able to accurately localize victims thus reducing response times and saving lives, (ii) building managers will be able to save up to 30% energy and money by conditioning each room in real-time based on the room's occupancy, (iii) people with disabilities will be able to use the technology for assistance such as finding wheelchair-accessible routes, and (iv) warehouse managers will be able to reduce order fulfillment time. Indoor localization will be, in the near future, as pervasive as GPS is today.