While GPS is the de-facto solution for outdoor positioning with a clear sky view, there is no prevailing technology for GPS-deprived areas, including dense city centers, urban canyons, buildings and other covered structures, and subterranean facilities such as underground mines, where GPS signals are severely attenuated or totally blocked. As an alternative to GPS for the outdoors, indoor localization using machine learning is an emerging embedded and Internet of Things (IoT) application domain that is poised to reinvent the way we navigate in various indoor environments. This book discusses advances in the applications of machine learning that enable the localization and navigation of humans, robots, and vehicles in GPS-deficient environments. The book explores key challenges in the domain, such as mobile device resource limitations, device heterogeneity, environmental uncertainties, wireless signal variations, and security vulnerabilities. Countering these challenges can improve the accuracy, reliability, predictability, and energy-efficiency of indoor localization and navigation. The book identifies severalnovel energy-efficient, real-time, and robust indoor localization techniques that utilize emerging deep machine learning and statistical techniques to address the challenges for indoor localization and navigation. In particular, the book: Provides comprehensive coverage of the application of machine learning to the domain of indoor localization; Presents techniques to adapt and optimize machine learning models for fast, energy-efficient indoor localization; Covers design and deployment of indoor localization frameworks on mobile, IoT, and embedded devices in real conditions.

This book provides a comprehensive and in-depth understanding of wireless indoor localization for ubiquitous applications. The past decade has witnessed a flourishing of WiFi-based indoor localization, which has become one of the most popular localization solutions and has attracted considerable attention from both the academic and industrial communities. Specifically focusing on WiFi fingerprint based localization via crowdsourcing, the book follows a top-down approach and explores the three most important aspects of wireless indoor localization: deployment, maintenance, and service accuracy. After extensively reviewing the state-of-the-art literature, it highlights the latest advances in crowdsourcing-enabled WiFi localization. It elaborated the ideas, methods and systems for implementing the crowdsourcing approach for fingerprint-based localization. By tackling the problems such as: deployment costs of fingerprint database construction, maintenance overhead of fingerprint database updating, floor plan generation, and location errors, the book offers a valuable reference guide for technicians and practitioners in the field of location-based services. As the first of its kind, introducing readers to WiFi-based localization from a crowdsourcing perspective, it will greatly benefit and appeal to scientists and researchers in mobile and ubiquitous computing and related areas.

Despite the enormous technical progress seen in the past few years, the maturity of indoor localization technologies has not yet reached the level of GNSS solutions. The 23 selected papers in this book present the recent advances and new developments in indoor localization systems and technologies, propose novel or improved methods with increased performance, provide insight into various aspects of quality control, and also introduce some unorthodox positioning methods.

In recent years, rapid development in robotics, mobile, and communication technologies has encouraged many studies in the field of localization and navigation in indoor environments. An accurate localization system that can operate in an indoor environment has considerable practical value, because it can be built into autonomous mobile systems or a personal navigation system on a smartphone for guiding people through airports, shopping malls, museums and other public institutions, etc. Such a system would be particularly useful for blind people. Modern smartphones are equipped with numerous sensors (such as inertial sensors, cameras, and barometers) and communication modules (such as WiFi, Bluetooth, NFC, LTE/5G, and UWB capabilities), which enable the implementation of various localization algorithms, namely, visual localization, inertial navigation system, and radio localization. For the mapping of indoor environments and localization of autonomous mobile sysems, LIDAR sensors are also frequently used in addition to smartphone sensors. Visual localization and inertial navigation systems are sensitive to external disturbances; therefore, sensor fusion approaches can be used for the implementation of robust localization algorithms. These have to be optimized in order to be computationally efficient, which is essential for realtime processing and low energy consumption on a smartphone or robot.

During the last century, navigation systems have become ubiquitous and guide drivers, cyclists, and pedestrians towards their desired destinations. While operating worldwide, they rely on line-of-sight conditions towards satellites and are thus limited to outdoor areas. However, finding a gate within an airport, a ward within a hospital, or a university's auditorium also represent navigation problems. To provide navigation within such indoor environments, new approaches are required. This thesis examines pedestrian 3D indoor localization and navigation using commodity smartphones: A desirable target platform, always at hand and equipped with a multitude of sensors. The IMU (accelerometer, gyroscope, magnetometer) and barometer allow for pedestrian dead reckoning, that is, estimating relative location changes. Absolute whereabouts can be determined via Wi-Fi, an infrastructure present within most public buildings, or by using Bluetooth Low Energy Beacons as inexpensive supplement. The building's 3D floorplan not only enables navigation, but also increases accuracy by preventing impossible movements, and serves as a visual reference for the pedestrian. All aforementioned information is fused by recursive density estimation based on a particle filter. The conducted experiments cover both, theoretical backgrounds and real-world use-cases. All discussed approaches utilize the infrastructure found within most public buildings, are easy to set up, and maintain. Overall, this thesis results in an indoor localization and navigation system that can be easily deployed, without requiring any special hardware components.

This book presents recent advances towards the goal of enabling efficient implementation of machine learning models on resource-constrained systems, covering different application domains. The focus is on presenting interesting and new use cases of applying machine learning to innovative application domains, exploring the efficient hardware design of efficient machine learning accelerators, memory optimization techniques, illustrating model compression and neural architecture search techniques for energy-efficient and fast execution on resource-constrained hardware platforms, and understanding hardware-software codesign techniques for achieving even greater energy, reliability, and performance benefits. Discusses efficient implementation of machine learning in embedded, CPS, IoT, and edge computing; Offers comprehensive coverage of hardware design, software design, and hardware/software co-design and co-optimization; Describes real applications to demonstrate how embedded, CPS, IoT, and edge applications benefit from machine learning.

Through expanded intelligence, the use of robotics has fundamentally transformed a variety of fields, including manufacturing, aerospace, medicine, social services, and agriculture. Continued research on robotic design is critical to solving various dynamic obstacles individuals, enterprises, and humanity at large face on a daily basis. Robotic Systems: Concepts, Methodologies, Tools, and Applications is a vital reference source that delves into the current issues, methodologies, and trends relating to advanced robotic technology in the modern world. Highlighting a range of topics such as mechatronics, cybernetics, and human-computer interaction, this multi-volume book is ideally designed for robotics engineers, mechanical engineers, robotics technicians, operators, software engineers, designers, programmers, industry professionals, researchers, students, academicians, and computer practitioners seeking current research on developing innovative ideas for intelligent and autonomous robotics systems.

The Present and Future of Indoor Navigation provides a complete overview of the latest indoor navigation technologies, algorithms, and systems. It begins by discussing various types of sensors that can be used for indoor navigation, such as accelerometers, gyroscopes, barometers, magnetometers, and cameras. It covers the numerous algorithms that can be used to compute the navigation solution, including Kalman filtering, particle filtering, and machine learning. Also, it discusses the system implementation considerations for indoor navigation, such as infrastructure, data fusion, and security. The book’s focus is on present technologies and algorithms, as well as providing a look into the future possibilities for indoor navigation, making it a great resource for a wide audience. This includes researchers, engineers, and students who are interested in indoor navigation. It is also a valuable resource for anyone who wants to learn more about the latest technologies and algorithms for indoor navigation.