Comments on “Adaptive variable structure set-point control ofunderactuated robots”

In this note, we point out the fallacies of above mentioned paper by Su et al. (ibid. vol.44 (1999)) concerning the stabilization of underactuated robots     

Adaptive variable structure set-point control of underactuatedrobots

Control of underactuated mechanical systems (robots) represents an important class of control problem. In this correspondence, a model-based adaptive variable structure control scheme is introduced, where the uncertainty bounds only depend on the inertia parameters of the system. Global asymptotic stability is established in the Lyapunov sense. Numerical simulations are conducted to validate the theoretical analysis     

Comments on “An improved Razumikhin-type theorem and itsapplications” [and reply]

Xu and Liu (see ibid., vol.39, no.4, p.839-41, 1994) tried to improve the well-known Razumikhin theorem on the stability of differential functional equations. The commenter states that, the proof of their main result contains an error, so the results may not be correct. In reply Bugong Xu gives a corrected result     

Comments on “Direct learning of control efforts fortrajectories with different time scales” [and reply]

In the original paper by Xu (ibid., vol.43, p.1027-30, 1998), a problem of learning motion trajectories with different time scales based on the learned control inputs was considered. It is suggested that the author was apparently unaware that such an idea was previously reported by Kawamura et al. (1987, 1995) based on the time-scale property of robot dynamics. In this note, we briefly review the literature and point out the difference between Xu's work and that of Kawamura et al. We show that the proof developed in the latter can also be applied to derive the time-scale learning controller for the class of nonlinear system considered in Xu. In reply, Xu acknowledges that he overlooked the work of Kawamura et al. He remarks that his original article was in the event published as 2 articles. He also mentions his work currently in progress     

Comments on “Nonparametric segmentation of curves intovarious representations” [and reply]

The author points out the existence of a theoretical difficulty that underlies the curve segmentation problem studied in the work of Rosin and West (ibid., vol.17, no.12, p.1140-53, 1995) and present a possible solution to it. The problem arises because curves are segmented into low- and high-order features, and the former are a subset of the latter. In reply, Rosin and West discuss the advantages and disadvantages of two approaches to model selection: their original heuristic sequential method, and the information theoretic method suggested by Kanatani and others     

Comments on “A new approach to fuzzy-neural systemmodeling” [and reply]

In the original paper (Y. Lin and G. A. Cunningham, ibid., vol. 3, p. 190-8, 1995), an approach to fuzzy-neural knowledge extraction starting from multi-input single-output examples is given. The author shows that the performance index given there is almost monotonically decreasing with m^-(1/2); that is, it is possible to obtain a very small performance index simply by increasing m. This is possible even if at the end of the learning phase the root mean square error remains very high. The original authors acknowledge that this is correct and explain why nevertheless they presented the approach that they did     

Comments on “The unreasonable effectiveness of neural networkapproximation” [and reply]

This paper discusses the misleading nature of the rate of approximation result reported in the paper by Dingankar (ibid. vol.44, 1999). In reply, Dingankar explains why the original statement of the theorem in that paper is correct as it stands, and clarifies the statements in the original paper. He points out how a proper understanding of the terminology resolves the first objection of the commentator and briefly outlines the directions he is pursuing in a separate research based on the second and the third recommendations     

Comments on “The effects of quantization on multilayer neuralnetworks” [and reply]

The commenters point out and correct the errors in the derivations of the equations appeared in the above paper by Dundar and Rose (ibid., vol.6 (1995)). In reply, Dundar and Rose agree with their corrections and state that the errors are relatively small and should not affect the conclusions.     

Comments on “Robust stability of linear systems with delayedperturbations” [and reply]

This paper comments on the result of the paper by Kim (ibid. vol.41 (1996)). The estimate of the stability robustness of linear time-delay systems in that paper is compared with the one which is solved in a Riccati matrix inequality framework. In reply, Kim acknowledge that his result needs the selection of a matrix Q and a scalar α which maximize the bound η_k0(Q, α). The result proposed by Ooba-Funahashi, that does not need this selection, is summarized     

Comments on “Design of supervised classifiers using Booleanneural networks” [and reply]

Gazula and Kabuka (1995) describe a binary neural network which implements a nonparametric statistical classifier. However, they implement a kernel-based classifier rather than k-nearest-neighbors, as stated in their paper. The commenter states that some other aspects of their paper are not clear. One of the original authors replies to points made by the commenter     

Comments on “Choquet fuzzy integral-based hierarchicalnetworks for decision analysis” [and reply]

We remark on an error in the above paper by Chiang (ibid. vol.7 (1999)). The purpose of this note is to present the correct formulas for partial derivatives of fuzzy integral-based neural nodes with respect to densities of Sugeno measures. In reply, Chiang agree with the correction     

Comments on “Geodesic saliency of watershed contours andhierarchical segmentation” [and reply]

In a paper on morphological image segmentation, Najman and Schmitt (1996) introduce the powerful concept of edge dynamics. In this communication, we show that the method that they propose to compute the edge dynamics gives erroneous results for certain spatial configurations, and we propose a new algorithm which always yields correct edge dynamics. The reply presents in detail the algorithm of the watershed, which have been sketched in the original article and criticized in the comment. First, the formal definition of the flooding list, the key data structure of the algorithm, is given. Then, the construction of this flooding list and of the watershed are described and proved     

Comments on “Factors that impact implementing a systemdevelopment methodology” [and reply]

In this correspondence, we point out some of the major shortcomings that we have discovered in the article titled “Factors that Impact Implementing a System Development Methodology” by Roberts et al. (1998). The article was published in the August 1998 issue of IEEE Transactions of Software Engineering. The article contains multiple problems that, if not pointed out, have the potential to lead to a state of confusion among researchers and practitioners alike. In particular, the article has the following problems: (1) The authors claim that the lack of theoretical basis for their factors is due to the fact that SDM implementation has never been studied in the literature. In fact, there is a multitude of studies very similar to that of Roberts at al. (2) The study does not meet commonly accepted standards for factor analysis procedures such as a minimum sample size ratio of 5:1, a ratio mandated even by the authors' own citations. This and other factor analysis problems lead to results that are questionable. In order to make the published article more useful to researchers and practitioners, this paper corrects some of the inaccuracies in the Roberts et al.'s article by: providing a brief literature review of some articles that are similar to the Roberts et al., study; and noting some of the technical inaccuracies in the data analysis procedures used by the authors so that the results can be interpreted in the proper context     

Comments regarding “On stability of interval matrices”[with reply]

We note that several of the results on Schur-stability of interval matrices in the above paper by Sezer-Siljak (ibid. vol.39) can be viewed as special cases of results obtained in a more general framework, and some of the latter were previously reported in the literature. In reply, Sezer-Siljak response to the issues raised in the comment     

Comments on “Analysis and design of fuzzy controller andfuzzy observer” [and reply]

The main advantage of Takagi-Sugeno (TS) fuzzy model is the ability to approximate the nonlinear systems from the input-output data without the complex mathematical equations. In order to obtain TS fuzzy model from the input-output data, all the system variables should be measurable. Therefore, in case of TS fuzzy model-based control, the observer design is of no good except the very special case. In this correspondence, we show this fact by analyzing the problems of the above paper by Ma et al. (ibid. vol.6 (1998)). In reply, Ma et al. point out that: 1) for TS fuzzy model, it is not necessary to make all the system variables measurable; 2) no accuracy is needed for the fuzzy model; and 3) the nonlinearities of the plant (31) given in the paper are the functions of variables x₁, x₂ and x₄     

Comments on “Vector norms as Lyapunov functions for linearsystems” [with reply]

The purpose of this note is to give a counterexample to a possible implication of the main result of the above paper by Kiendl-Adamy-Stelzner (ibid., vol.37, no.6, p.839-42 (1992)), to modify this result, to restate its theorem accordingly, and to supply a proof for it. In reply, Kiendl et al. point out that they did not formulate the condition as necessary for a vector norm being a Lyapunov function. Due to Hmamed's remarks, they have reviewed their statements for imprecise formulations     

Comments on “Combinatorial rule explosion eliminated by afuzzy rule configuration” [and reply]

In the original paper (IEEE Trans. Fuzzy Syst., vol.6, p.1-11, 1998), Combs and Andrews proved the following logical equivalence (stated here for two antecedents p and q and one consequent r, but easily generalize to an arbitrary number of antecedents and consequents): [(p∧q)⇒r]⇔[(p⇒r)V(q⇒r)]. This is a very significant result because it suggests that we can replace multi-antecedent rules with an interconnection of single antecedent rules, which eliminates the rule explosion that is associated with multi-antecedent rules. Combs and Andrews refer to the left-hand side of this equivalence as an intersection rule configuration (IRC) and to its right-hand side as a union rule configuration (URC). Their result gives rise to two distinctly different paths for the design of fuzzy logic systems; IRC, which leads to rule explosion, and URC, which does not. The authors discuss four points about the IRC⇔URC relation. The original authors reply, acknowledging some of the points and stating that they would present their results differently if starting now     

Comments on “Robust, fragile, or optimal?” [and reply]

The issue of fragile controllers produced by using popular robust and optimal control synthesis methods was raised by Keel-Bhattacharyya in the above mentioned paper (ibid. vol.42 (1997)). The commenter points out that the controller fragility problem described in the paper is: 1) well known in the literature; 2) due to the poor choice of examples; 3) can easily be fixed by changing the optimization criteria; and 4) can be fixed by changing the implementation. In reply, Keel-Bhattacharyya disagree with him on all counts and explain their argument     

Comments on “Some conditions on zeros to avoid step-responseextrema”. II [and reply]

The commenters give a correct version of Theorem 2 of the above-mentioned paper (Rachid, ibid., vol.40, p.1501-3, 1995). In reply Rachid accepts one of their points and gives a new statement of Theorem 2