The Pontis bridge management system is the predominant bridge management system employed in the United States. The system employs a network optimization model for preservation, formulated as a Markov decision process with a linear program solution procedure. On each bridge, a set of level-of-service standards determines functional needs, whose benefits are calculated according to a user cost model. A multi-year program simulation generates project alternatives by combining preservation and improvement needs on each bridge. The program is optimized within budget constraints by means of an incremental benefit/cost algorithm.The mathematical formulation of each of these components is presented and discussed. Aspects of system development and data management are outlined, along with the current implementation status. California's experience with the use of Pontis in its funding process is highlighted. ed approach of optimizing the economic efficiency of the bridge network. In today's complex political and financial environment, the system provides a defensible and understandable means of expressing the long-term benefits of keeping bridges in good condition, as well as an objective way of choosing among maintenance, improvement. and replacement opportunities. In order to succeed in this objective. Pontis has the capacity to express the engineering concerns of deterioration and structural performance in economic terms understandable to a broader audience.
As an enhancement of the technology championed in VR, cinematic rendering promises to provide additional anatomic detail for MDCT interpretation and display. Future investigations must be conducted to evaluate the diagnostic accuracy of cinematic rendering and determine whether interpretative pitfalls result from its unique lighting model in practice.
Markovian bridge deterioration models have been in use in the United States since the early 1990s, using the AASHTO Guide to Commonly-Recognized Structural Elements. California has one of the oldest databases of inspection history using this standard. A preliminary analysis was performed on the California data set, to quantify the deterioration transition probabilities actually observed, and to determine whether it is yet possible to validate the key assumptions of Markovian bridge deterioration models. Several important conclusions were reached, which should provide guidance for future research and implementation.
After the discovery of a significant crack in an eyebar (fracture-critical element) on the San Francisco–Oakland Bay Bridge in California, the California Department of Transportation (Caltrans) explored possible remote monitor solutions to provide the greatest possible safeguards for the some 200,000 vehicles that used the bridge daily. Caltrans selected an acoustic emission (AE) monitoring system that allowed the detection and localization of crack initiation and growth in real time. The AE method relies on the propagation of elastic waves released by a sudden stress–strain change at the crack tip. The challenge of the AE method in a field application is the disturbance of the data by extraneous noise sources. A robust damage detection algorithm is required to differentiate relevant damage data from secondary noise sources, such as friction. Caltrans required that the proposed monitoring solution be validated through laboratory testing. This paper presents the full-scale laboratory testing of two eyebars loaded under constant amplitude fatigue to detect and locate crack growth under intense friction-type noise sources with use of the AE method. Verification is presented of AE results with ultrasonics testing, dye penetrant, and visual methods to detect damage. Linear location accuracy with simulated signal sources is discussed. The pattern recognition algorithm to differentiate noise signals and crack growth signals is presented. On the basis of this study, the AE method will be used to continuously monitor 384 eyebars with 640 AE sensors on the San Francisco–Oakland Bay Bridge.
As virtual worlds become increasingly complex, task level interaction with virtual actors becomes correspondingly important. The control problem simply becomes unmanageable if we try to interact with synthetic agents at the wrong level of abstraction. However, it is not sufficient merely to implement a set of behaviours for a virtual actor; we require some mechanism for selecting and sequencing motor skills appropriate to the current behavioural goals and the states of other objects and actors in the virtual environment. In this paper we will describe a mechanism for linking perception and action to generate routine behaviours in a process we call motor planning. We present our implementation of the skill network, in which motor skills are the nodes and the arcs represent inhibitory and excitatory connections, including extensions to this architecture based on recent work in robotics. We characterize the domain of motor planning, i.e. what kinds of behaviour can it account for, and when will it fail? We close with a discussion of the limits of our current implementation and work that remains to be done.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.