Showing papers on "Autonomous system (mathematics) published in 1992"

Prediction of chaotic time series with neural networks and the issue of dynamic modeling

Abstract: This paper deals with the role of neural-network based prediction for the modeling of nonlinear dynamical systems. We show experimentally that the backpropagation learning rule to train neural networks and the prediction error, so widely utilized in teaching and comparing nonlinear predictors, do not consistently indicate that the neural network based model has indeed captured the dynamics of the system that produced the time series. Frequently, but not always, the neural network when used as an autonomous system in a feedback configuration was able to generate a time series that has dynamical invariants similar to the original time series. We show that the estimation of the dynamical invariants (correlation dimension, largest Lyapunov exponent) of the predicted and original time series are an appropriate tool to validate the predictive model.

Autonomous processing in parallel distributed processing networks

Abstract: This paper critically examines the claim that parallel distributed processing (PDP) networks are autonomous learning systems. A PDP model of a simple distributed associative memory is considered. It is shown that the ‘generic’ PDP architecture cannot implement the computations required by this memory system without the aid of external control. In other words, the model is not autonomous. Two specific problems are highlighted: (i) simultaneous learning and recall are not permitted to occur as would be required of an autonomous system; (ii) connections between processing units cannot simultaneously represent current and previous network activation as would be required if learning is to occur. Similar problems exist for more sophisticated networks constructed from the generic PDP architecture. We argue that this is because these models are not adequately constrained by the properties of the functional architecture assumed by PDP modelers. It is also argued that without such constraints, PDP research...

Explicit Path Tracking by Autonomous Vehicles

Abstract: We have suggested a novel approach to autonomously navigate a full sized autonomous vehicle that separately treats vehicle control and obstacle detection. In this paper we discuss the vehicle control that has enabled our autonomous vehicle to travel at speeds upto 20mph. We point out the limitations of existing schemes that restrict their consideration to kinematic models and show that it is possible to obtain an increase in performance through the use of approximate dynamical models that capture first–order effects. Our approach combines such a modeling philosophy with accurate feedback in world coordinates from sensors that have only recently become available. Experimental results of our implementation on NavLab, a modified van at CMU, are presented.

Autonomous system expansion planning considering renewable energy sources-a computer package

Abstract: The authors describe the interactive package ASP (Autonomous System Planning), which determines the optimal expansion plan of an autonomous generation system including diesel units, wind generators, and photovoltaic generators. Utilizing meteorological data, load demand data and estimated economical parameters, the package finds the optimal system expansion policy for a period of N years. The solution algorithms implemented in the package recognize planning and operational constraints and take into consideration the stochastic nature of the meteorological conditions, loads, and the availability of the diesel units. >

A Real-time Algorithm for Obstacle Avoidance of Autonomous Mobile Robots

Abstract: A real-time obstacle avoidance algorithm is proposed for autonomous mobile robots. The algorithm is sensor-based and consists of a H-mode and T-mode. The algorithm can deal with a complicated obstacle environment, such as multiple concave and convex obstacles. It will be shown that the algorithm is more efficient and more robust than other sensor-based algorithms. In addition, the algorithm will guarantee a solution for the obstacle avoidance problem. Since the algorithm only takes up a small computational time, it can be implemented in real time.

Autonomous robotic disassembly in the blocks world

Abstract: This article describes a fully functional autonomous system with two cooperating robots that disassembles complex Duplo1structures in a restricted environment. The system operates on a table-top world with any number of Duplo structures for which object models have been given. The structures are dis assembled down to individual parts, using basic operations of single part removal, object partitioning, and the addition of stabilizing supports to structures that would otherwise fall over. All aspects are automatically planned, including operation selection, path and grasp planning, and simultaneous cooper ative robot motion. Operations are chosen so as to not create unstable structures and to not risk breakage in areas of low structural integrity. Overall planning is done with a process mechanism that heuristically generates efficient disassembly sequences, without searching a space of all possible operations. Several examples of actual system operation, using real robots, are presented.

Injecting inter-autonomous system routes into intra-autonomous system routing: a performance analysis

Abstract: The current TCP/IP Internet may be modeled as an arbitrary interconnection of autonomous systems 1 ( [Honig 90]) . Routing in the Interne t is partitioned into intra-autonomous system routing (intra.-S routing) an d inter-autonomous system routing (inter-AS routing) . If both a source and a destination belong to the same AS . then routing between the source and th e destination is completely realized by employing intra-AS routing procedure s used by the AS . However . if a source and a destination belong to differen t ASs . then routing between the source and the destination cannot be realized either completely within the inter-AS routing or completely within th e intra-AS routing . but involves both the intra and inter-AS routing . That . i n turn, requires a certain degree of interaction between the intra and inter-A S routing procedures .

U. S. Air Force Phillips Laboratory Autonomous Space Navigation Experiment

Abstract: The Air Force Phillips Laboratory’s Technology for Autonomous Operational Survivability (TAOS) space experiment is scheduled for launch on an Air Force Space Test Program Space Test Experiment Platform (STEP) spacecraft in early 1993. The mission will test and evaluate two navigation systems that support autonomous satellite navigation. They are the Microcosm Autonomous Navigation System (MANS) and the Rockwell Autonetics Six-Channel Global Positioning System (GPS) Receiver. MANS is a true autonomous system that uses horizon scanners, modified for Sun and Moon detection, as primary measurement devices to determine position, velocity, and attitude and estimate position and velocity using a Kalman filter. MANS supplemental sensors include GPS and an inertial measurement unit (IMU). The Rockwell miniature GPS receiver is a semi-autonomous system which accesses the GPS network to determine spacecraft position, velocity, and time. Position and velocity reference data will be generated using direct measurements from Air Force Satellite Control Network (AFSCN) remote tracking stations (RTS), orbit reconstruction based on on-board beaconry, and post-processed GPS data and solutions. The attitude reference will be provided by an on-board strapdown IMU.

Preventive control in an autonomous decentralized system

Abstract: For more rational operation and control of future electric power systems which would continue to expand in size and complexity, an autonomous decentralized control system is proposed as one of the new control architectures. In the autonomous decentralized control system, each generating station and substation behaves independently and cooperatively without being controlled by the central station. Besides the inherent fault-tolerant characteristics of the architecture, flexible and rapid control depending on the circumstances can be achieved. In this paper, the fundamental logic of preventive control in an autonomous decentralized system is developed. It is based on the bidding method in which each station proposes its own output taking into consideration its own situation and cooperation with others. The economic load dispatch including preventive control can be performed, based on the bidding rule adopting the equivalent incremental generating cost curve in which the security constraints are incorporated. Application of the logic to a model system revealed satisfactory control performance for preventing the overload of transmission lines and the undervoltage.

Path keeping of autonomous underwater vehicles using sliding mode control

Abstract: Path keeping of autonomous underwater vehicles is formulated as a nonlinear state space control problem subject to disturbances, modelling errors, and parameter uncertainty. The system to be controlled is nonlinear with parameters that vary with speed and operational conditions. Accurate path keeping autopilot functions are necessary for the autonomous navigation of a vehicle through confined spaces, and in the presence of obstacles and cross-current environments. The above uncertainties in the force coefficients and environmental disturbances, as well as the required accuracy lead to the need for a robust control for successful vehicle operations. Recent developments of variable structure control laws in the form of sliding modes have been shown to provide added robustness that is quite remarkable for autopilot designs. This paper investigates the use of a cross track error guidance law with a sliding mode compensator and presents results based on computer simulations using a nonlinear dynamic model of an existing vehicle. The results illustrate that the influence of speed, modelling nonlinearity and uncertainty, as well as disturbances, can be effectively compensated, leading to path keeping accuracy.

On exact and approximate modeling of exponential time series

Abstract: The authors consider the problem of identifying linear time-invariant models from frequency-response measurements. They depart from the standard setting of this problem by refraining from a priori assumptions on the input-output structure of the to-be-identified system. Without this structure, the optimal model which explains the exponential time series is characterized by an autonomous system. By a posteriori imposing the structure of input and output variables of the system, it is shown that the set of optimal (i.e., least complexity) input-output models can be derived in a straightforward way from a polynomial representation of the optimal autonomous system. The authors formulate this as a minimal complexity covering problem and provide a conceptual algorithm for the computation of its solution. The problem of approximate modeling of purely exponential data is also discussed. An equation error type of misfit function is considered, and a class of approximate input-output models which satisfy a uniform upper bound of the misfit criterion is characterized. >

Graphical Analysis: The Phase Plane

Abstract: This chapter discusses the procedure and application of the phase plane. It is applicable to two coupled autonomous first-order ordinary differential equations or an autonomous second-order ordinary differential equation. It yields a graphical representation of the solution. The qualitative features of the solution of two coupled autonomous first-order ordinary differential equations may be ascertained from the phase plane. The set of two coupled autonomous first-order ordinary differential equations is assumed. There exist systems in which the arrows are pointing in the opposite direction than what has been illustrated. Each solution describes a single trajectory. Every trajectory must either go to infinity, or approach a limit cycle.