Perceiving,learning, and exploiting object affordances for autonomous pile manipulation

Autonomous manipulation in unstructured environments will enable a large variety of exciting and important applications. Despite its promise, autonomous manipulation remains largely unsolved. Even the most rudimentary manipulation task—such as removing objects from a pile—remains challenging for robots. We identify three major challenges that must be addressed to enable autonomous manipulation: object segmentation, action selection, and motion generation. These challenges become more pronounced when unknown man-made or natural objects are cluttered together in a pile. We present a system capable of manipulating unknown objects in such an environment. Our robot is tasked with clearing a table by removing objects from a pile and placing them into a bin. To that end, we address the three aforementioned challenges. Our robot perceives the environment with an RGB-D sensor, segmenting the pile into object hypotheses using non-parametric surface models. Our system then computes the affordances of each object, and selects the best affordance and its associated action to execute. Finally, our robot instantiates the proper compliant motion primitive to safely execute the desired action. For efficient and reliable action selection, we developed a framework for supervised learning of manipulation expertise. To verify the performance of our system, we conducted dozens of trials and report on several hours of experiments involving more than 1,500 interactions. The results show that our learning-based approach for pile manipulation outperforms a common sense heuristic as well as a random strategy, and is on par with human action selection.     

A memetic algorithm for the capacitated m-ring-star problem

The Capacitated m-Ring-Star Problem (CmRSP) models a network topology design problem in the telecommunications industry. In this paper, we propose to solve this problem using a memetic algorithm that includes a crossover operation, a mutation operation, a local search involving three neighborhood operators, and a population selection strategy that maintains population diversity. Our approach generates the best known solutions for 131 out of 138 benchmark instances, improving on the previous best solutions for 24 of them, and exhibits more advantages on large benchmark instances when compared with the best existing approach. Additionally, all existing algorithms for this problem in literature assume that the underlying graph of the problem instance satisfies the triangle inequality rule; our approach does not require this assumption. We also generated a new set of 36 larger test instances based on a digital data service network price structure to serve as a new benchmark data set for future researchers.     

Methanol oxidation on Au/TiO2 catalysts

We have investigated the adsorption and reaction of methanol with Au/TiO₂ catalysts using a pulsed flow reactor, DRIFTS and TPD. The TiO₂ (P25) surface adsorbed a full monolayer of methanol, much of it in a dissociative manner, forming methoxy groups associated with the cationic sites, and hydroxyl groups at the anions. The methoxy is relatively stable until 250 °C, at which point decomposition occurs, producing mainly dimethyl ether by a bimolecular surface reaction. As the concentration of methoxy on the surface diminishes, so the mechanism reverts to a de-oxygenation pathway, producing mainly methane and water (at ~330 °C in TPD), but also with some coincident CO and hydrogen. Au catalysts were prepared by the deposition-precipitation method to give Au loadings between 0.5–3 wt %. The effect of low levels of Au on the reactivity is marked. The pathway which gives methane, which is characteristic of titania, remains, but a new feature of the reaction is the evolution of CO₂ and H₂ at lower temperature (a peak is seen in TPD at 220 °C), and the elimination of the DME-producing state. Clearly this is associated with the presence of Au and appears to be due to the production of a formate species on the surface of the Au component. This formate species is mainly involved in the reaction of methanol with the Au/TiO₂ catalysts which results in a combustion pathway being followed, with complete conversion occurring by ~130 °C.     

Feeding response of Ips paraconfusus to phloem and phloem metabolites of Heterobasidion annosum-inoculated ponderosa pine,Pinus ponderosa

In studies of feeding by the bark beetle, Ips paraconfusus, two pine stilbenes (pinosylvin and pinosylvin methyl ether), ferulic acid glucoside, and enantiomers of the four most common sugars present in ponderosa pine phloem (sucrose, glucose, fructose, and raffinose) did not stimulate or reduce male feeding when assayed on wet -cellulose with or without stimulatory phloem extractives present. When allowed to feed on wet -cellulose containing sequential extracts (hexane, methanol, and water) of ponderosa pine phloem, methanol and water extractives stimulated feeding, but hexane extractives did not. Males confined in wet -cellulose containing aqueous or organic extracts of culture broths derived from phloem tissue and containing the root pathogen, Heterobasidion annosum, ingested less substrate than beetles confined to control preparations. In an assay using logs from uninoculated ponderosa pines, the mean lengths of phloem in the digestive tracts increased as time spent feeding increased. Males confined to the phloem of basal logs cut from ponderosa pines artificially inoculated with H. annosum ingested significantly less phloem than beetles in logs cut from trees that were (combined) mock-inoculated or uninoculated and did not contain the pathogen. However, individual pathogen-containing treatments were not significantly different from uninoculated controls. It was concluded that altered feeding rates are not a major factor which may explain why diseased ponderosa pines are colonized by I. paraconfusus.     

Lipid hydrolysis and oxidation on the surface of milled rice

The changes in milled rice FFA content and composition and in conjugated diene (CD) content and bacterial, yeast, and mold counts were determined at 24, 37, and 50°C and 70% RH over 50 d. There was a rapid rate of FFA formation during the first few days of storage, which was optimal at 37°C, but that slowed after 2, 4, and 5 d at 37, 24, and 50°C, respectively. There was a second increase in FFA after about day 12 that increased with increasing temperature, indicating nonlipase hydrolysis. Linoleic and oleic acids were the main components of the total FFA produced on the surface of milled rice. The pattern of CD development followed that of FFA increase. Bacterial growth correlated with increased FFA levels after 12 d of storage, suggesting that bacterial lipase rather than bran lipase may be responsible for rice lipid hydrolysis     

In situ observation of methoxide hydrogenation on the Ni(111) surface

The surface chemistry of methoxide (CH₃O-) on the Ni(111) surface has been studied in the presence of hydrogen pressures up to 2 Torr. During heating in vacuum methoxide decomposes to H₂ and CO, which desorb at 380 and 445 K, respectively. The CH₃O-decomposition process is rate limited by CH bond breaking and exhibits a strong deuterium kinetic isotope effect in CD₃O-. In the presence of ambient hydrogen pressures of 0.02–2.0 Torr both CH₃O- and CD₃O-are hydrogenated directly to methanol at 310 K. Methoxide is hydrogenated by adsorbed hydrogen, which nearly saturates the surface at these pressures and temperatures.     

Comparing techniques for authorship attribution of source code

Attributing authorship of documents with unknown creators has been studied extensively for natural language text such as essays and literature, but less so for non‐natural languages such as computer source code. Previous attempts at attributing authorship of source code can be categorised by two attributes: the software features used for the classification, either strings of n tokens/bytes (n‐grams) or software metrics; and the classification technique that exploits those features, either information retrieval ranking or machine learning. The results of existing studies, however, are not directly comparable as all use different test beds and evaluation methodologies, making it difficult to assess which approach is superior. This paper summarises all previous techniques to source code authorship attribution, implements feature sets that are motivated by the literature, and applies information retrieval ranking methods or machine classifiers for each approach. Importantly, all approaches are tested on identical collections from varying programming languages and author types. Our conclusions are as follows: (i) ranking and machine classifier approaches are around 90% and 85% accurate, respectively, for a one‐in‐10 classification problem; (ii) the byte‐level n‐gram approach is best used with different parameters to those previously published; (iii) neural networks and support vector machines were found to be the most accurate machine classifiers of the eight evaluated; (iv) use of n‐gram features in combination with machine classifiers shows promise, but there are scalability problems that still must be overcome; and (v) approaches based on information retrieval techniques are currently more accurate than approaches based on machine learning. Copyright © 2012 John Wiley & Sons, Ltd.     

Alkali silica reactivity of agglomerated silica fume

It is commonly accepted that replacement of a portion of cement in mortar or concrete with well-dispersed silica fume reduces expansion caused by alkali silica reaction. Recently there has been much discussion that large, agglomerated particles of silica fume may actually act as alkali silica reactive aggregates, thereby increasing expansion rather than reducing it. The data in the literature, from both field and laboratory studies, are inconsistent. This prompted an extensive laboratory investigation into the alkali silica reactivity of silica fume. Results from accelerated expansion testing and microscopic investigations are presented. It was seen that some agglomerated silica fumes participate in ASR while others do not. Factors determining the reactivity of silica fume agglomerates are suggested.     

Length preferences and periodicity in {beta}-strands. Antiparallel edge {beta}-sheets are more likely to finish in non-hydrogen bonded rings

We analysed the length distributions of different types of ß-strandin a high resolution, non-homologous set of 500 protein structures,finding differences in their mean lengths. Antiparallel edgestrands in strand–turn–strand motifs show a preferencefor an even number of residues. This propensity is enhancedif the length is corrected for ß-bulges, which insertan extra residue into the strand. Residues in antiparallel edgeß-strands alternate between being in hydrogen bondedand non-hydrogen bonded rings. Antiparallel edges with an evennumber of residues are more likely to have their final ßresidue in a non-hydrogen bonded ring. This suggests that non-hydrogenbonded rings are intrinsically more stable than hydrogen bondedrings, perhaps because its side chain packing is closer. Therefore,we suggest that a simple way to increase ß-hairpinstability, or the stability of an antiparallel edge strand,is to have a non-hydrogen bonded ring at the end of the strand. Received June 19, 2003; revised October 25, 2003; accepted November 7, 2003     

Antimicrobial activity of statistical polymethacrylic sulfopropylbetaines against gram‐positive and gram‐negative bacteria

A series of statistical copolymers derived from 2‐(dimethylamino)ethyl methacrylate with four different hydrophobic comonomers (ethyl, butyl, cyclohexyl, and octyl methacrylates) have been prepared via conventional free radical copolymerization under bulk conditions. The copolymers have been subsequently modified, with 1,3‐propanesultone to yield the corresponding polysulfoproylbetaine derivatives. Those copolymers exhibiting the requisite aqueous solubility have been screened with respect to their antimicrobial activity against two common and notorious pathogens, namely Staphylococcus aureus (S. aureus) and Escherichia coli (E. coli). We show that certain copolymers do indeed exhibit antimicrobial activity. The extent of activity is related to the molecular characteristics of the materials such as the molar composition and structure of the hydrophobic comonomer. © 2006 Wiley Periodicals, Inc. J Appl Polym Sci 101: 1036–1041, 2006