The growth characteristics, morphology, and biochemical activity of four strains of a hemolytic, urease-positive, gram-negative organism isolated from vaginal exudate of postparturient sows closely resembled those of organisms belonging to the genus Actinobacillus. Comparison by agglutination, immunodiffusion, and electrophoresis in acrylamide gel revealed that the four strains were identical. They were distinguished from A . lignieresii, A . equuli, A . seminis, A . suis, and other members of the family Brucellaceae by means of these three techniques. The four strains were related antigenically t o all strains of the genus Actinobacillus examined, but were related most closely t o A . seminis and A . suis. The antigenic relatedness and other similarities support inclusion of the organism in the genus Actinobacillus. Its biochemical, antigenic, and electrophoretic differences from established species of the genus Actinobacillus support the conclusion that this swine actinobacillus is a new species. A species name is not proposed for this organism because of the present uncertain taxonomic status of related actinobacilli. The swine actinobacillus persisted in the vagina of two sows for at least 40 days after intravaginal inoculation; however, proof that vaginal infection with the organism causes urogenital disorders of any type has not been obtained. Intraperitoneal and intravenous inoculation of four 6-week-old colostrum-deprived pigs with the organism resulted in no clinical or postmortem evidence of disease. Strain 192 (= ATCC 27072 = NCTC 10801) is the representative strain of this group of swine actinobacilli.

No abstract

This paper considers the problem of designing a class of robust algorithms for the trajectory tracking control of an uncertain robot manipulator. The general control structure consists of two parts: the primary control law is first introduced to stabilize the nominal system and then a class of robust non-linear control laws are adopted to compensate for the system uncertainties by using the deterministic approach. The uncertainties assumed are bounded by higher-order polynomials in the Euclidean norms of system states with known (or unknown) bounding coefficients. The possible bounds of uncertainties are assumed to be known for the robust non-linear control with less computatiomal burden. If no information on these bounds is available, then the adaptive bound of the robust controller is presented to overcome possible time-varyin uncertainties, that is a decentralized adaptive control scheme. With a feasible class of desired trajectories, the proposed controf laws guarantee that all possible responses of the corresponding closed-loop ,systems are at least uniformly ultimately bounded (UUB) by Lyapunov stability theory. The effectiveness of the proposed control algorithms are verified through numerical simulations. Finally, it is shown that the presented controllers are evaluated to be robust with respect to a given class of uncertainties.Unless mentioned otherwise, the following standard notations and terminology will be used in the sequel.CP E n f t ) : R""" + R" inf LP R R + a positive (negative) definite matrix A the induced matrix norm of a real matrix A E R""" corresponding to the Euclidean vector norm; ((Al( = { a-(ATA)} the closed ball in R" of radius r > 0 centred at x = 0: B,(x): = (X E R": 1 1~1 1 < r)the set of p-times continuously differentiable functions an (n x n) identity matrix f is a function, mapping a domain R" " into the range R" the infimum, the greatest lower bound the function norm in the Lebesgue space. Let f ( t ) : R + R" be the Lebesgue measurable function and f

No abstract

No abstract