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.     

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.     

Designing for video: investigating the contextual cues within viewing situations

The viewing of video increasingly occurs in a wide range of public and private environments via a range of static and mobile devices. The proliferation of content on demand and the diversity of the viewing situations means that delivery systems can play a key role in introducing audiences to contextually relevant content of interest whilst maximising the viewing experience for individual viewers. However, for video delivery systems to do this, they need to take into account the diversity of the situations where video is consumed, and the differing viewing experiences that users desire to create within them. This requires an ability to identify different contextual viewing situations as perceived by users. This paper presents the results from a detailed, multi-method, user-centred field study with 11 UK-based users of video-based content. Following a review of the literature (to identify viewing situations of interest on which to focus), data collection was conducted comprising observation, diaries, interviews and self-captured video. Insights were gained into whether and how users choose to engage with content in different public and private spaces. The results identified and validated a set of contextual cues that characterise distinctive viewing situations. Four archetypical viewing situations were identified: ‘quality time’, ‘opportunistic planning’, ‘sharing space but not content’ and ‘opportunistic self-indulgence’. These can be differentiated in terms of key contextual factors: solitary/shared experiences, public/private spaces and temporal characteristics. The presence of clear contextual cues provides the opportunity for video delivery systems to better tailor content and format to the viewing situation or additionally augment video services through social media in order to provide specific experiences sensitive to both temporal and physical contexts.     

Synthetic hydrogels: 3. Hydroxyalkyl acrylate and methacrylate copolymers: surface and mechanical properties

The surface and mechanical properties of copolymers of hydroxyalkyl acrylates and methacrylates have been examined by a variety of techniques. This work is complementary to earlier parts of this series which describe the effect of copolymer structure on water binding properties. Water structure has been demonstrated to exert a profound effect upon mechanical properties whether measured in compression or in tension. In particular, water that is characterized by differential scanning calorimetry as ‘freezing’ water is observed to have a marked plasticizing effect upon the gel, whereas ‘non-freezing’ water has little such effect. Similarly, the ‘freezing’ water produces a more marked effect on thermally induced transitions. Two distinct transition points are observed as a result of its presence. One corresponds to the freezing point of water and the other to a glass transition temperature, whose value depends upon the proportion of ‘freezing’ or ‘plasticizing’ water in the gel. Several predictive and direct measurement techniques have been used to study the surface properties of the copolymers in both hydrated and dehydrated states. Taken together they have established a sound understanding of the way in which polar and dispersive components of surface free energy vary as a function of copolymer composition and water content. Use of protein adsorption and fibroblast cell interaction techniques demonstrate that biological phenomena respond to changes at a molecular level which current macroscopic surface energy techniques are unable to discern.     

Activated gas plasma surface treatment of polymers for adhesive bonding

Polyethylene, polypropylene, poly(vinyl fluoride) (Tedlar), polystyrene, nylon 6, poly(ethylene terephthalate) (Mylar), polycarbonate, cellulose acetate butyrate, and a poly(oxymethylene) copolymer were treated with activated helium and with activated oxygen. Mechanical strengths of adhesive-bonded specimens prepared from treated and from untreated coupons were compared. Polyethylene (PE) and polypropylene (PP) showed the greatest increases in bond strength. Oxygen and helium were both effective with polyethylene, but polypropylene showed no improvement when treated with activated helium. The results with excited helium parallel the effects of ionizing radiation on these two polymers, as does the appearance of unsaturation bands in the infrared (965 cm^?1 in PE, and 887 and 910 cm^?1 in PP). Active nitrogen produced excellent bond strength with polyethylene but not with polypropylene. Of the remaining polymers examined, Tedlar, polystyrene, and nylon 6 showed the greatest improvement in bondability after treatment, and Mylar showed moderate improvement. Polycarbonate, cellulose acetate butyrate, and the poly(oxymethylene) copolymer gave approximately two-fold increases in lap-shear bond strength. In several cases, significant differences in response to time of treatment and type of excited gas were found.     

Prevention of water pollution problems in mining: The bactericide technology

In pyritic environments, the bacteriaThiobacillus ferrooxidans catalyze acid formation by increasing the oxidation rate of pyrite by a factor of one million. This acid solubilizes metals and pollutes adjacent streams and lands. Bactericide sprays during mining and waste disposal operations attack the source of the problem by preventing acid formation and metals solubilization. Used in conjunction with current water treatment systems, bactericides can dramatically reduce operating costs. Controlled release bactericides contribute to successful reclamation by providing assurance against revegetation failure and post-reclamation water quality problems that can necessitate perpetual water treatment. While inhibitingT. ferrooxidans these organic compounds aid in the establishment of beneficial heterotrophic bacteria which support vegetation. These conditions continue to persist after the bactericide is depleted from the controlled release system. Case Studies I and II show that bactericides inhibit acid generation during hard rock and coal mining operations and they are cost effective. Case Studies III and IV illustrate the improvement in water quality and vegetation after reclamation when controlled release bactericides were used. Economic analyses show cost benefits are achieved when controlled-release bactericides are part of the reclamation plan.     

Object tracking based on an online learning network with total error rate minimization

This paper presents a visual object tracking system which is tolerant to external imaging factors such as illumination, scale, rotation, occlusion and background changes. Specifically, an integration of an online version of total-error-rate minimization based projection network with an observation model of particle filter is proposed to effectively distinguish between the target object and the background. A re-weighting technique is proposed to stabilize the sampling of particle filter for stochastic propagation. For self-adaptation, an automatic updating scheme and extraction of training samples are proposed to adjust to system changes online. Our qualitative and quantitative experiments on 16 public video sequences show convincing performances in terms of tracking accuracy and computational efficiency over competing state-of-the-art algorithms.     

PalmPhasor算法性能的理论分析

模板的安全性和隐私性是掌纹系统实际应用的关键问题,然而生物特征保护的多项指标通常相互冲突并且难以同时满足.作为解决上述冲突的一种可撤销掌纹编码算法,PalmPhasor实现了高效、安全的掌纹认证.建立了系统分析PalmPhasor性能的完整框架.为了便于具体分析,将情景分为4种情况,并且提供了支持相应分析的预备知识,包括辅助定理以及Gabor滤波掌纹图像实部和虚部分布特性.在统计学基础上建立的理论分析和实验结果均表明:即使在用户口令被盗的情况下,多方向分数级融合增强的PalmPhasor算法也可以同时有效地满足可撤销生物特征的4项指标.