Self-organizing approach to the stochastic fuel regulator problem
Stochastic control optimization, examining self organizing approach for performance adaptive controller with asymptotic optimal properties
Gilbert, H. D.↗
SEARCH · Search NASA
Results for “Stochastic optimization”
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Mathematical model of a flexible space shuttle vehicle
The development of a mathematical model of the lateral motion of a flexible space shuttle vehicle during ascent is described. The model was developed to perform control system synthesis using stochastic constrained optimization techniques. The goals of the control system synthesis are to demonstrate the applicability of the techniques and to discover any problems peculiar to the flexible nature of a shuttle vehicle. The equations of motion are derived. A brief description of the generation of numerical data is given. Explicit definitions and numerical values of trajectory data and coefficients appearing in the equations of motion are included.
Harvey, C. A.↗
Evaluation of performance characteristics for a space antenna system subjected to stochastic disturbances
Control system synthesis for a large space antenna system is investigated based on stochastic linear optimal control techniques and the minimization of a quadratic Gaussian performance index. Parametric studies indicate that suitable combinations of plant and sensor noise characteristics, and state weighting matrices, can be found to meet the mission RMS pointing requirements. Using a combination of the Kalman filter and linear feedback, an optimal control law for the finite element model of the Hoop/Column structural system without damping is obtained. Removing the hoop-mounted actuator resulted in an increase in the RMS errors, an increased control effort, and an increase in the least damped modal time constant. Removing the hoop-mounted sensor resulted in an RMS error increase and estimator performance degradation, with less system RMS performance degradation than noted for removal of the hoop-mounted actuator.
Ananthakrishnan, S.↗
Randomly Sampled-Data Control Systems
The purpose is to solve the Linear Quadratic Regulator (LQR) problem with random time sampling. Such a sampling scheme may arise from imperfect instrumentation as in the case of sampling jitter. It can also model the stochastic information exchange among decentralized controllers to name just a few. A practical suboptimal controller is proposed with the nice property of mean square stability. The proposed controller is suboptimal in the sense that the control structure is limited to be linear. Because of i. i. d. assumption, this does not seem unreasonable. Once the control structure is fixed, the stochastic discrete optimal control problem is transformed into an equivalent deterministic optimal control problem with dynamics described by the matrix difference equation. The N-horizon control problem is solved using the Lagrange's multiplier method. The infinite horizon control problem is formulated as a classical minimization problem. Assuming existence of solution to the minimization problem, the total system is shown to be mean square stable under certain observability conditions. Computer simulations are performed to illustrate these conditions.
Han, Kuoruey↗
Scheduling Earth Observing Satellites with Evolutionary Algorithms
We hypothesize that evolutionary algorithms can effectively schedule coordinated fleets of Earth observing satellites. The constraints are complex and the bottlenecks are not well understood, a condition where evolutionary algorithms are often effective. This is, in part, because evolutionary algorithms require only that one can represent solutions, modify solutions, and evaluate solution fitness. To test the hypothesis we have developed a representative set of problems, produced optimization software (in Java) to solve them, and run experiments comparing techniques. This paper presents initial results of a comparison of several evolutionary and other optimization techniques; namely the genetic algorithm, simulated annealing, squeaky wheel optimization, and stochastic hill climbing. We also compare separate satellite vs. integrated scheduling of a two satellite constellation. While the results are not definitive, tests to date suggest that simulated annealing is the best search technique and integrated scheduling is superior.
Globus, Al↗
A Novel LDPP-MADDPG Approach for Distributed Power Allocation in mmWave Cellular Networks
This paper considers the problem of distributed beam scheduling and power allocation problem in millimeter- Wave (mmWave) cellular networks, in which multiple Base Stations (BSs) operate as individual operators over a shared spectrum. We propose a novel learning-aided approach that integrates the Lyapunov Drift-Plus-Penalty (LDPP) framework and Multi-agent Deep Deterministic Policy Gradient (MADDPG) reinforcement learning algorithms. This offers a powerful approach to learning stable and constraint-aware policies, reaping the joint benefit of both LDPP and MADDPG, in complex multiagent environments. The major challenge for this approach is to integrate these two approaches in a meaningful and effective manner. The key idea to solve this problem is to introduce a novel feature of local observation that incorporates potential negative value of the reward function due to the stochastic constraints introduced by the LDPP framework. Empirical results demonstrate that our proposed scheme outperforms the baseline methods under various conditions.
99 - GENERAL AND MISCELLANEOUS↗
Design of optimal partial state feedback controllers for linear systems in stochastic environments
The problem of obtaining an optimal control law, which is constrained to be a feedback of the available measurements, is considered for both continuous and discrete time linear systems subjected to additive white process noise and measurement noise. Necessary conditions are obtained for minimizing a quadratic performance function for both finite and infinite duration cases. The feedback gain matrices are constrained to be constant for the infinite duration cases. For all the cases considered, algorithms are derived for generating sequences of feedback gain matrices which successively improve the performance function. Computational aspects are discussed via application to two continuous time processes, including a helicopter/slung load system subjected to measurement noise and random wind gust input.
Joshi, S. M.↗
The control of nonlinear stochastic control systems under discounted performance criteria
Control algorithm for optimization of nonlinear stochastic processes with discounted performance criteria
Harris, C. A.↗
Optimal Control via Self-Generated Stochasticity
The problem of global maxima of functionals has been examined. Mathematical roots of local maxima are the same as those for a much simpler problem of finding global maximum of a multi-dimensional function. The second problem is instability even if an optimal trajectory is found, there is no guarantee that it is stable. As a result, a fundamentally new approach is introduced to optimal control based upon two new ideas. The first idea is to represent the functional to be maximized as a limit of a probability density governed by the appropriately selected Liouville equation. Then, the corresponding ordinary differential equations (ODEs) become stochastic, and that sample of the solution that has the largest value will have the highest probability to appear in ODE simulation. The main advantages of the stochastic approach are that it is not sensitive to local maxima, the function to be maximized must be only integrable but not necessarily differentiable, and global equality and inequality constraints do not cause any significant obstacles. The second idea is to remove possible instability of the optimal solution by equipping the control system with a self-stabilizing device. The applications of the proposed methodology will optimize the performance of NASA spacecraft, as well as robot performance.
Zak, Michail↗
Preliminary study of auto-differentiation algorithm in beam dynamics with stochastic process
Modern particle accelerator optimization requires sophisticated computational methods to address the inherently stochastic nature of beam dynamics. This research develops a framework applying AD to SDEs that specifically addresses beam dynamics challenges in particle accelerators, focusing on accurately modeling and optimizing beam behavior in regimes dominated by stochastic processes. By incorporating key physical phenomena such as synchrotron radiation, wakefield effects, and quantum excitation, the framework aims to provide auto differentiation on the figure of merit of the phase space evolution and beam dynamics. The methodology will enable effective optimization method in a dynamic system with stochastic process.
Accelerator Physics↗
Optimization of Operations Resources via Discrete Event Simulation Modeling
The resource levels required for operation and support of reusable launch vehicles are typically defined through discrete event simulation modeling. Minimizing these resources constitutes an optimization problem involving discrete variables and simulation. Conventional approaches to solve such optimization problems involving integer valued decision variables are the pattern search and statistical methods. However, in a simulation environment that is characterized by search spaces of unknown topology and stochastic measures, these optimization approaches often prove inadequate. In this paper, we have explored the applicability of genetic algorithms to the simulation domain. Genetic algorithms provide a robust search strategy that does not require continuity and differentiability of the problem domain. The genetic algorithm successfully minimized the operation and support activities for a space vehicle, through a discrete event simulation model. The practical issues associated with simulation optimization, such as stochastic variables and constraints, were also taken into consideration.
Joshi, B.↗
Chance-Constrained Guidance With Non-Convex Constraints
Missions to small bodies, such as comets or asteroids, require autonomous guidance for descent to these small bodies. Such guidance is made challenging by uncertainty in the position and velocity of the spacecraft, as well as the uncertainty in the gravitational field around the small body. In addition, the requirement to avoid collision with the asteroid represents a non-convex constraint that means finding the optimal guidance trajectory, in general, is intractable. In this innovation, a new approach is proposed for chance-constrained optimal guidance with non-convex constraints. Chance-constrained guidance takes into account uncertainty so that the probability of collision is below a specified threshold. In this approach, a new bounding method has been developed to obtain a set of decomposed chance constraints that is a sufficient condition of the original chance constraint. The decomposition of the chance constraint enables its efficient evaluation, as well as the application of the branch and bound method. Branch and bound enables non-convex problems to be solved efficiently to global optimality. Considering the problem of finite-horizon robust optimal control of dynamic systems under Gaussian-distributed stochastic uncertainty, with state and control constraints, a discrete-time, continuous-state linear dynamics model is assumed. Gaussian-distributed stochastic uncertainty is a more natural model for exogenous disturbances such as wind gusts and turbulence than the previously studied set-bounded models. However, with stochastic uncertainty, it is often impossible to guarantee that state constraints are satisfied, because there is typically a non-zero probability of having a disturbance that is large enough to push the state out of the feasible region. An effective framework to address robustness with stochastic uncertainty is optimization with chance constraints. These require that the probability of violating the state constraints (i.e., the probability of failure) is below a user-specified bound known as the risk bound. An example problem is to drive a car to a destination as fast as possible while limiting the probability of an accident to 10(exp -7). This framework allows users to trade conservatism against performance by choosing the risk bound. The more risk the user accepts, the better performance they can expect.
FROM↗
Determine Optimal Stimulus Amplitude for Using Vestibular Stochastic Stimulation to Improve Balance Function
Sensorimotor changes such as postural and gait instabilities can affect the functional performance of astronauts when they transition across different gravity environments. We are developing a method, based on stochastic resonance (SR), to enhance information transfer by applying non-zero levels of external noise on the vestibular system (vestibular stochastic resonance, VSR). Our previous work has shown the advantageous effects of VSR in a balance task of standing on an unstable surface [1]. This technique to improve detection of vestibular signals uses a stimulus delivery system that provides imperceptibly low levels of white noise-based binaural bipolar electrical stimulation of the vestibular system. The goal of this project is to determine optimal levels of stimulation for SR applications by using a defined vestibular threshold of motion detection. A series of experiments were carried out to determine a robust paradigm to identify a vestibular threshold that can then be used to recommend optimal stimulation levels for sensorimotor adaptability (SA) training applications customized to each crewmember. The amplitude of stimulation to be used in the VSR application has varied across studies in the literature such as 60% of nociceptive stimulus thresholds [2]. We compared subjects' perceptual threshold with that obtained from two measures of body sway. Each test session was 463s long and consisted of several 15s long sinusoidal stimuli, at different current amplitudes (0-2 mA), interspersed with 20-20.5s periods of no stimulation. Subjects sat on a chair with their eyes closed and had to report their perception of motion through a joystick. A force plate underneath the chair recorded medio-lateral shear forces and roll moments. Comparison of threshold of motion detection obtained from joystick data versus body sway suggests that perceptual thresholds were significantly lower. In the balance task, subjects stood on an unstable surface and had to maintain balance, and the stimulation was administered from 20-400% of subjects' vestibular threshold. Optimal stimulation amplitude was determined at which the balance performance was best compared to control (no stimulation). Preliminary results show that, in general, using stimulation amplitudes at 40-60% of perceptual motion threshold significantly improved the balance performance. We hypothesize that VSR stimulation will act synergistically with SA training to improve adaptability by increasing utilization of vestibular information and therefore will help us to optimize and personalize a SA countermeasure prescription. This combination may help to significantly reduce the number of days required to recover functional performance to preflight levels after long-duration spaceflight.
Goel, R.↗
ProxSTORM - A Stochastic Trust Region Method for Nonsmooth Optimization
Explore the source record for details and available documents.
Javeed, Aurya↗
Absolute value optimization to estimate phase properties of stochastic time series
Most existing deconvolution techniques are incapable of determining phase properties of wavelets from time series data; to assure a unique solution, minimum phase is usually assumed. It is demonstrated, for moving average processes of order one, that deconvolution filtering using the absolute value norm provides an estimate of the wavelet shape that has the correct phase character when the random driving process is nonnormal. Numerical tests show that this result probably applies to more general processes.
Scargle, J. D.↗
Determine Optimal Stimulus Amplitude for Using Vestibular Stochastic Stimulation to Improve Balance Function
No abstract available
Goel, Rahul↗
Guidance and Control strategies for aerospace vehicles
A neighboring optimal guidance scheme was devised for a nonlinear dynamic system with stochastic inputs and perfect measurements as applicable to fuel optimal control of an aeroassisted orbital transfer vehicle. For the deterministic nonlinear dynamic system describing the atmospheric maneuver, a nominal trajectory was determined. Then, a neighboring, optimal guidance scheme was obtained for open loop and closed loop control configurations. Taking modelling uncertainties into account, a linear, stochastic, neighboring optimal guidance scheme was devised. Finally, the optimal trajectory was approximated as the sum of the deterministic nominal trajectory and the stochastic neighboring optimal solution. Numerical results are presented for a typical vehicle. A fuel-optimal control problem in aeroassisted noncoplanar orbital transfer is also addressed. The equations of motion for the atmospheric maneuver are nonlinear and the optimal (nominal) trajectory and control are obtained. In order to follow the nominal trajectory under actual conditions, a neighboring optimum guidance scheme is designed using linear quadratic regulator theory for onboard real-time implementation. One of the state variables is used as the independent variable in reference to the time. The weighting matrices in the performance index are chosen by a combination of a heuristic method and an optimal modal approach. The necessary feedback control law is obtained in order to minimize the deviations from the nominal conditions.
Hibey, J. L.↗
Tuning Monotonic Basin Hopping: Improving the Efficiency of Stochastic Search as Applied to Low-Thrust Trajectory Optimization
Trajectory optimization methods using monotonic basin hopping (MBH) have become well developed during the past decade [1, 2, 3, 4, 5, 6]. An essential component of MBH is a controlled random search through the multi-dimensional space of possible solutions. Historically, the randomness has been generated by drawing random variable (RV)s from a uniform probability distribution. Here, we investigate the generating the randomness by drawing the RVs from Cauchy and Pareto distributions, chosen because of their characteristic long tails. We demonstrate that using Cauchy distributions (as first suggested by J. Englander [3, 6]) significantly improves monotonic basin hopping (MBH) performance, and that Pareto distributions provide even greater improvements. Improved performance is defined in terms of efficiency and robustness. Efficiency is finding better solutions in less time. Robustness is efficiency that is undiminished by (a) the boundary conditions and internal constraints of the optimization problem being solved, and (b) by variations in the parameters of the probability distribution. Robustness is important for achieving performance improvements that are not problem specific. In this work we show that the performance improvements are the result of how these long-tailed distributions enable MBH to search the solution space faster and more thoroughly. In developing this explanation, we use the concepts of sub-diffusive, normally-diffusive, and super-diffusive random walks (RWs) originally developed in the field of statistical physics.
autonomous↗