No abstract

The aerospace industry has been adopting avionics architectures to take advantage of advances in computer engineering. Integrated Modular Avionics (IMA), as described in ARINC 653, distributes functional modules into a robust configuration interconnected with a "virtual backplane" data communications network. Each avionics module's function is defined in software compliant with the APEX Application Program Interface. The Avionics Full-Duplex Ethernet (AFDX) network replaces the point-topoint connections used in previous distributed systems with "virtual links". This network creates a command and data path between avionics modules with the software and network defining the active virtual links over an integrated physical network. In the event of failures, the software and network can perform complex reconfigurations very quickly, resulting in a very robust system.In this paper, suitable architectures, standards and conceptual designs for IMA computational modules and the virtual backplane are defined and analyzed for applicability to spacecraft. The AFDX network standard is examined in detail and compared with IEEE 802.3 Ethernet. A reference design for the "Ancillary Sensor Network" (ASN) is outlined based on the IEEE 1451 "Standard for a Smart Transducer Interface for Sensors and Actuators" using realtime operating systems, time deterministic AFDX and wireless LAN technology. Strategies for flight test and operational data collection related to Systems Health Management are developed, facilitating vehicle ground processing. Finally, a laboratory evaluation defines performance metrics and test protocols and summarizes the results of AFDX network tests, allowing identification of design issues and determination of ASN subsystem scalability, from a few to potentially thousands of smart and legacy sensors. 12

No abstract

A near-infrared (NIR) spectroscopic method has been developed to monitor flowing pharmaceutical powders during their voiding and detect post-blending segregation. The method is capable of providing both chemical and physical information (particle size differences) on the flowing pharmaceutical powders. Particle size differences are widely recognized as the predominant driver for segregation. Pharmaceutical powders may segregate following blending as they are voided down pipes to compressing machines, increasing the variability of the drug content and dissolution of the final product tablets because of segregation. NIR diffuse reflectance spectra of pharmaceutical powders were obtained following voiding through a six-foot pipe. Spectral subtraction was used to eliminate baseline differences but maintain particle size differences. The NIR spectra indicated differences in the particle size of the flowing powder. Particle size differences were also tracked throughout the voiding of pharmaceutical powders by plotting the absorbance at 1536 nm. The method was also applied to the voiding of two layers of lactose particles with different particle sizes. The system described in this report provides an approach to study post-blending segregation in pharmaceutical powders and other relevant materials.

Many industrial applications demand accurate ranging operations, for example, robotic endeffector positioning and measurement of vibrations in machine monitoring tasks. For instance, measurement of the position of a robotic end effector in high-accuracy applications demands the following stringent specifications from a ranging system: ( 1 ) absolute ranging accuracy of +_ 0.025 mm ( q-0.001 in. ), (2) working range of approximately 2 m (6.56 ft), (3) update rate > 100 Hz, and (4) high reliability. It is shown that an ultrasonic ranging system that meets the above specifications may be feasible. The system model is based on measuring the distance between a transmitter and a receiver. A novel method to detect the instant when the signal arrives at the receiver is described. The method combines envelope peak detection and phase measurement to determine the time of arrival, and it gets close to meeting the accuracy criteria. To verify the analytical predictions, distances of up to 0.9 m were measured using a prototype system. The absolute accuracy of such measurements was q-0.150 mm ( q-0.006 in. ). The tests performed indicate that the errors are caused primarily by air turbulence and may be decreased by averaging more samples. In this paper ultrasonic ranging from a transmitter to a receiver in a three-dimensional space setting is addressed.