Le développement de comportements utiles pour les robots complexes, tel que des humanoïdes, s'avère difficile. La commande corps-complet à base de modèle allège en partie ces difficultés, en permettant la composition des comportements corps-complets complexes à partir de plusieurs tâches atomiques effectuées simultanément sur le robot. Cependant, des hypothèses et erreurs de modélisation, faites pendant la planification des tâches, peuvent produire des combinaisons infaisables/incompatibles quand exécutées sur le robot, créant des mouvements corps-complet imprévisibles, et probablement dangereux. L'objectif de ce travail est de mieux comprendre ce qui rend les tâches infaisables ou incompatibles et de développer des méthodes automatiques pour améliorer ces problèmes pour que les mouvements corps-complets puissent être accomplis comme prévu. Nous commençons par construire un formalisme permettant d'analyser quand les tâches sont faisables et compatibles étant données les contraintes de commande. En utilisant les métriques de faisabilité et compatibilité à base de modèle, nous démontrons comment optimiser les tâches avec des outils de commande prédictive non-linéaire ainsi que les inconvénients de cette approche. Afin de surmonter ces faiblesses, une boucle d'optimisation est formulée, qui améliore automatiquement la faisabilité et compatibilité des tâches via la recherche de politique sans modèle en conjonction avec la commande corps-complets à base de modèle. À travers une série d'expériences simulées et réelles, nous montrons que la simple optimisation de faisabilité et compatibilité des tâches nous permet de réaliser des mouvements corps-complets utiles. / Producing useful behaviors on complex robots, such as humanoids, is a challenging undertaking. Model-based whole-body control alleviates some of this difficulty by allowing complex whole-body motions to be broken up into multiple atomic tasks, which are performed simultaneously on the robot. However, modeling errors and assumptions, made during task planning, often result in infeasible and/or incompatible task combinations when executed on the robot. Consequently, there is no guarantee that the prescribed tasks will be accomplished, resulting in unpredictable, and most likely, unsafe whole-body motions. The objective of this work is to better understand what makes tasks infeasible or incompatible, and develop automatic methods of improving on these two issues so that the overall whole-body motions may be accomplished as planned. We start by building a concrete analytical formalism of what it means for tasks to be feasible with the control constraints and compatible with one another. Using the model-based feasibility and compatibility metrics, we demonstrate how the tasks can be optimized using non-linear model predictive control, while also detailing the shortcomings of this model-based approach. In order to overcome these weaknesses, an optimization loop is designed, which automatically improves task feasibility and compatibility using model-free policy search in conjunction with model-based whole-body control. Through a series of simulated and real-world experiments, we demonstrate that by simply optimizing the tasks to improve both feasibility and compatibility, complex and useful whole-body motions can be realized.

The aim of diploma theses is the design of an adaptive robotic workplace. The theoretical part focus on the division of robotic systems and the technical description of individual devices that were used in the implementation of the solution. The practical part contains an analysis of solutions and optimization of the entire production process in terms of minimizing the trajectory, smoothness of movements, time interval, which were analyzed in RoboSim software and in Roboshop software source code was created. Part of the theses is the design for an adaptive production process. The result of the work is an algorithm for controlling robot movements between individual processes. The theses contain a variant solution and possible innovative solutions for possible expansion of the workplace.

In this work, capturing Near-Earth Asteroids (NEAs) into Near-Earth orbits is investigated. A general optimization strategy is employed whereby a genetic algorithm is used to seed a sequential quadratic programming (SQP) method for the first step, and then nearby solutions seed further SQP runs. A large number of solutions are produced for several asteroids with varying levels of thrust, and under the effects of various perturbations. Solutions are found over a range of epochs and times of flight as opposed to many traditional methods of optimizing point solutions. This methodology proved effective, finding low-thrust capture solutions within 10% of the required Delta V for analytically estimated transfers, and matching results from other optimization programs such as MALTO. It is found that the effects of solar radiation pressure (SRP) and n-body effects do not have a significant impact on the optimized transfer costs, nor do the perturbations significantly affect the shapes and trends of the optimized solution space. <p><br></p> <p>These optimized results are then used to develop analytic models for both optimized transfer costs and flight times. These models are then used to estimate the transfer costs and flight times for all listed Near Earth Asteroids from the JPL Small Body Database. This analysis is then used to determine the nominal properties of potentially capturable asteroids. The characteristics are then related to a series of different asteroid transfer technologies, elucidating each technology's capabilities and potential capture targets. Finally, this analysis concludes with a brief roadmap of the major decisions mission designers should consider for future asteroid capture missions.</p>

This paper studies the impacts on flight trajectories, such as lateral and vertical changes, when avoiding the formation of persistent contrails for transatlantic flights. A sophisticated Earth-System Model (EMAC) coupled with a flight routing submodel (AirTraf) and a contrail submodel (CONTRAIL) is used to optimize flight trajectories concerning the flight time and the flight distance through contrail forming regions (contrail distance). All the trajectories are calculated taking into account the effects of the actual and local meteorological parameters, e.g., wind, temperature, relative humidity, etc. A full-year simulation has been conducted based on a daily flight schedule of 103 transatlantic flights. The trade-off between the flight time and contrail distance shows a large daily variability, meaning for the same increase in flight time, the reduction in contrail distance varies from 20% to 80% depending on the daily meteorological situation. The results confirm that the overall changes in flight trajectories follow a seasonal cycle corresponding to the nature of the potential contrail coverage. In non-summer seasons, the southward and upward shifts of the trajectories are favorable to avoid the contrail formation. In summer, the northward and upward shifts are preferred. A partial mitigation strategy for up to 40% reduction in contrail distance can be achieved throughout all the seasons with a negligible increase in flight time (less than 2%), which represents a reasonable trade-off between flight time increase and contrail avoidance.

Intelligent transportation systems involve a variety of information and control systems methodologies, from cooperative systems which aim at traffic flow optimization by means of vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communication, to information fusion from multiple traffic sensing modalities. This thesis aims to address three problems in intelligent transportation systems, one in optimal control of connected automated vehicles, one in discrete-event and hybrid traffic simulation model, and one in sensing and classifying roadway obstacles in smart cities. The first set of problems addressed relates to optimally controlling connected automated vehicles (CAVs) crossing an urban intersection without any explicit traffic signaling. A decentralized optimal control framework is established whereby, under proper coordination among CAVs, each CAV can jointly minimize its energy consumption and travel time subject to hard safety constraints. A closed-form analytical solution is derived while taking speed, control, and safety constraints into consideration. The analytical solution of each such problem, when it exists, yields the optimal CAV acceleration/deceleration. The framework is capable of accommodating for turns and ensures the absence of collisions. In the meantime, a measurement of passenger comfort is taken into account while the vehicles make turns. In addition to the first-in-first-out (FIFO) ordering structure, the concept of dynamic resequencing is introduced which aims at further increasing the traffic throughput. This thesis also studies the impact of CAVs and shows the benefit that can be achieved by incorporating CAVs to conventional traffic. To validate the effectiveness of the proposed solution, a discrete-event and hybrid simulation framework based on SimEvents is proposed, which facilitates safety and performance evaluation of an intelligent transportation system. The traffic simulation model enables traffic study at the microscopic level, including new control algorithms for CAVs under different traffic scenarios, the event-driven aspects of transportation systems, and the effects of communication delays. The framework spans multiple toolboxes including MATLAB, Simulink, and SimEvents. In another direction, an unsupervised anomaly detection system is developed based on data collected through the Street Bump smartphone application. The system, which is built based on signal processing techniques and the concept of information entropy, is capable of generating a prioritized list of roadway obstacles, such that the higher-ranked entries are most likely to be actionable bumps (e.g., potholes) requiring immediate attention, while those lower-ranked are most likely to be nonactionable bumps(e.g., flat castings, cobblestone streets, speed bumps) for which no immediate action is needed. This system enables the City to efficiently prioritize repairs. Results on an actual data set provided by the City of Boston illustrate the feasibility and effectiveness of the system in practice.

Constrained optimization is used to derive human postures and movements. In the first study a static 3D model with 30 muscle groups is used to analyse postures. The activation levels of these muscles are minimized in order to represent the individual's choice of posture. Subject specific data in terms of anthropometry, strength and orthopedic aids serve as input. The aim is to study effects from orthopedic treatment and altered abilities of the subject. Initial validation shows qualitative agreement of posture strategies but further details about passive stiffness and anthropometry are needed, especially to predict pelvis orientation. In the second application, the athletic long jump, a problem formulation is developed to find optimal movements of a multibody system when subjected to contact. The model was based on rigid links, joint actuators and a wobbling mass. The contact to the ground was modelled as a spring-damper system with tuned properties. The movement in the degrees of freedom representing physical joints was described over contact time through two fifth-order polynomials, with a variable transition time, while the motion in the degrees of freedom of contact and wobbling mass was integrated forwards in time, as a consequence. Muscle activation variables were then optimized in order to maximize ballistic flight distance. The optimization determined contact time, end configuration, activation and interaction with the ground from an initial configuration. The results from optimization show a reasonable agreement with experimentally recorded jumps, but individual recordings and measurements are needed for more precise conclusions.

Connected aviation, the Internet of Flying Things and related emerging technologies, such as the System-Wide Information Management infrastructure of the FAA NextGen program, present numerous opportunities for the aviation sector. The ubiquity of aeronautical, flight, weather, aerodrome, and maintenance data accelerates the development of smarter software systems to cope with the ever increasing requirements of the industry sector. The increasing amount, frequency and variety of real-time data available to modern air transport and tactical systems, and their crews, creates exciting new challenges and research opportunities. We present an architectural approach toward the vision of increasingly self-separating and self-governed flight operations within the bigger picture of an evolving set of future Autonomous Flight Rules. The challenges in this field of research are manifold and include autonomic airborne trajectory optimization, data sharing, fusion and information derivation, the incorporation of and communication with rational actors—both human and machine—via a connected aviation infrastructure, to facilitate smarter decision making and support while generating economical, environmental and tactical advantages. We developed a concept and prototype implementation of our Smart Autoflight Control System. The concept and implemented system follow the design principle of an Autonomic Element, consisting of an Autonomic Manager and its Managed Element, acting within an Autonomic Context. The Managed Element concept embraces an infrastructure featuring suitable models of manageable environments, airborne agents, planners, applicable operational cost and risk policies, and connections to the System-Wide Information Management cloud as well as to relevant rational actors, such as Air Traffic Control, Command and Control, Operations or Dispatch. The Autonomic Manager concept incorporates the extraction, that is, short-term sensing, of features from operational scenarios and the categorization of these scenarios according to their level of criticality and associated flight phase. The Autonomic Manager component, furthermore, continuously tunes, that is, actuates, manageable items of its Managed Element, such as environments and planners, and triggers competitions to assess their performance under the various extracted and dynamically changing features of their Autonomic Context. The performance reputations of the tuned manageable items are collected in a knowledge base and may serve as a long-term sensor. Both the managed items of the Managed Element as well the managing items of the Autonomic Manager are extendable and may realize very different paradigms, including deterministic, non-deterministic, heuristically guided, and biologically inspired approaches. We assessed the extensibility and maintainability of our Smart Autoflight Control System infrastructure by including manageable environments and planners of the Classical Grid Search, Probabilistic Roadmaps, and Rapidly-Exploring Random Trees families into its core component. Furthermore, we evaluated the viability of a simple heuristic and a more sophisticated Sequential Model-Based Algorithm Configuration Autonomic Manager to adaptively select and tune manageable planners of the supported families based on the extracted features from very simple to highly challenging scenarios. We were able to show that a self-adaptive approach, that heuristically tunes and selects the best performing planner following a performance competition, produces suitable flight trajectories within reasonable deliberation times. Additionally, we discovered options for improving our heuristic Autonomic Manager through a series of evaluation runs of the Sequential Model-Based Algorithm Configuration Autonomic Manager. Our contributions answer how the manageable items, that is, environments and planners, of our Smart Autoflight Control System core component have to be modified in order to embed System-Wide Information Management data that feature both spatial and temporal aspects. We show how operational cost and risk policies help to assess environments differently and plan suitable flight trajectories accordingly. We identify and implement the necessary extensions and capabilities that have to be supported by manageable and managing items, respectively, to enable continuous feature extraction, adaptive tuning, performance competitions, and planner selection in dynamic flight scenarios. / Graduate

No description available.

No description available.

No description available.