No abstract

No abstract

A generalised registration technique based on mixed-mode measurements from two or more spatially distributed sensors i:j presented. The algorithm outputs include maximum likelihood estimates of (i) registration parameters, (ii) registered sensor measurements and (Si) registered target location estimates expressed in a common coordinate system. Numerical results are presented for two sensors with fixed but unknown measurement biases.

No abstract

Outlier-contaminated normal errors in regression problems are modelled by exponential power distributions and the resulting maximum likelihood estimators are shown to involve Lp minimisations (1 < P ,,2). It is shown that La estimation is minimax outlier-robust and minimax covariance-robust over the neighbourhood of exponential power distributions. Efficiency loss is negligible. Recursive gradient-type Lp estimators are derived and shown to be convergent and consistent. The major limitation on outlier robustness is seen to be the requirement for convergence of the recursive minimisation. The algorithm is validated with an application in adaptive control.