In recent years, the subject of physical interaction for aerial robots has been a popular research area with many new mechanical designs and control approaches being proposed. The aerial robotics community is currently observing a paradigm shift from classic guidance, navigation, and control tasks towards more unusual tasks, for example requesting aerial robots to physically interact with the environment, thus extending the manipulation task from the ground into the air. This thesis contributes to the field of aerial manipulation by proposing a novel concept known has Multiple Aerial-Ground Manipulator System or MAGMaS, including what appears to be the first experimental demonstration of a MAGMaS and opening a new route of research. The motivation behind associating ground and aerial robots for cooperative manipulation is to leverage their respective particularities, ground robots bring strength while aerial robots widen the workspace of the system. The first contribution of this work introduces a meticulous system model for MAGMaS. The system model's properties and potential extensions are discussed in this work. The planning, estimation and control methods which are necessary to exploit MAGMaS in a cooperative manipulation tasks are derived. This works proposes an optimal control allocation scheme to exploit the MAGMaS redundancies and a general model-based force estimation method is presented. All of the proposed techniques reported in this thesis are integrated in a global architecture used for simulations and experimental validation. This architecture is extended by the addition of a tele-presence framework to allow remote operations of MAGMaS. The global architecture is validated by robust demonstrations of bar lifting, an application that gives an outlook of the prospective use of the proposed concept of MAGMaS. Another contribution in the development of MAGMaS consists of an exploratory study on the flexibility of manipulated loads. A vibration model is derived and exploited to showcase vibration properties in terms of control. The last contribution of this thesis consists of an exploratory study on the use of elastic joints in aerial robots, endowing these systems with mechanical compliance and energy storage capabilities. Theoretical groundings are associated with a nonlinear controller synthesis. The proposed approach is validated by experimental work which relies on the integration of a lightweight variable stiffness actuator on an aerial robot.

Aerial robots, meaning robots with flying capabilities, are essentially robotic platforms, which are autonomously controlled via some sophisticated control engineering tools. Similar to aerial vehichles, they can overcome the gravitational forces thanks to their design and/or actuation type. What makes them different from the conventional aerial vehicles, is the level of their autonomy. Reducing the complexity for piloting of such robots/vehicles provide the human operator more freedom and comfort. With their increasing autonomy, they can perform many complicated tasks by their own (such as surveillance, monitoring, or inspection), leaving the human operator the most high-level decisions to be made, if necessary. In this way they can be operated in hazardous and challenging environments, which might posses high risks to the human health. Thanks to their wide range of usage, the ongoing researches on aerial robots is expected to have an increasing impact on the human life. Aerial Physical Interaction (APhI) is a case, in which the aerial robot exerts meaningful forces and torques (wrench) to its environment while preserving its stable flight. In this case, the robot does not try avoiding every obstacle in its environment, but prepare itself for embracing the effect of a physical interaction, furthermore turn this interaction into some meaningful robotic tasks. Aerial manipulation can be considered as a subset of APhI, where the flying robot is designed and controlled in purpose of manipulating its environment. A clear motivation of using aerial robots for physical interaction, is to benefit their great workspace and agility. Moreover, developing robots that can perform not only APhI but also aerial manipulation can bring the great workspace of the flying robots together with the vast dexterity of the manipulating arms. This thesis work is addressing the design, modeling and control problem of these aerial robots for the purpose of physical interaction and manipulation. Using the nonlinear mathematical models of the robots at hand, in this thesis several different control methods (IDA-PBC, Exact Linearization, Differential Flatness Based Control) for APhI and aerial manipulation tasks have been developed and proposed. Furthermore, novel design tools (e.g. new rigid/elastic manipulating arms, hardware, software) to be used together with miniature aerial robots are presented within this thesis, which contributes to the robotics society not only in terms of concrete theory but also practical implementation and experimental robotics.

In this dissertation, we focus on two fundamental problems related to the navigation of ground robots and small Unmanned Aerial Vehicle (UAVs): cooperative localization and path planning. The theme running through in all of the work is the use of bearing only sensors, with a focus on monocular video cameras mounted on ground robots and UAVs.

Unmanned aerial vehicles are increasingly being used to perform complex functions or to assist humans to carry out dangerous missions within dynamic environments. Other possible applications include search and rescue, disaster relief operations, environmental monitoring, wireless surveillance networks, and cooperative manipulation. Creating these types of autonomous aerial vehicles places severe demands on the design of control schemes that can adapt to different scenarios and possible changes of vehicle dynamics. In this book we address the challenging problem of employing aerial robots to transport and manipulate loads safely and efficiently. Aerial load manipulation and transportation is extremely important in emergency rescue missions as well as for military and industrial purposes. This book gives an insight into problems that can arise in aerial load transportation and suggests control systems techniques to solve them. A key focus is given on modeling of the aerial load transportation system as well as stability and robustness analysis. A detailed design and derivation of control algorithms based on adaptive control, optimal control and reinforcement learning are discussed in detail. Furthermore, an experimental testbed and controller implementation are delineated.

The Chemical Batch Reactor is aimed at tackling the above problems from a blending of academic and industrial perspectives. Advanced solutions (i.e., those based on recent research results) to the four fundamental problems of modeling, identification, control and fault diagnosis for batch processes are developed in detail in four distinct chapters. In each chapter, a general overview of foundational concepts is also given, together with a review of recent and classical literature on the various subjects. To provide a unitary treatment of the different topics and give a firm link to the underlying practical applications, a single case study is developed as the book progresses; a batch process of industrial interest, i.e., the phenol-formaldehyde reaction for the production of phenolic resins, is adopted to test the various techniques developed. In this way, a roadmap of the solutions to fundamental problems, ranging from the early stages of the production process to the complete design of control and diagnosis systems, is provided for both industrial practitioners and academic researchers.

Presents a unified treatment of HRI-related issues, identifies key themes, and discusses challenge problems that are likely to shape the field in the near future. The survey includes research results from a cross section of the universities, government efforts, industry labs, and countries that contribute to HRI.

This Proceedings Volume documents recent cutting-edge developments in multi-robot systems research and is the result of the Second International Workshop on Multi-Robot Systems that was held in March 2003 at the Naval Research Laboratory in Washington, D.C. This Workshop brought together top researchers working in areas relevant to designing teams of autonomous vehicles, including robots and unmanned ground, air, surface, and undersea vehicles. The workshop focused on the challenging issues of team architectures, vehicle learning and adaptation, heterogeneous group control and cooperation, task selection, dynamic autonomy, mixed initiative, and human and robot team interaction. A broad range of applications of this technology are presented in this volume, including UCAVS (Unmanned Combat Air Vehicles), micro-air vehicles, UUVs (Unmanned Underwater Vehicles), UGVs (Unmanned Ground Vehicles), planetary exploration, assembly in space, clean-up, and urban search and rescue. This Proceedings Volume represents the contributions of the top researchers in this field and serves as a valuable tool for professionals in this interdisciplinary field.