Compliant skills acquisition and multi-optima policy search with EM-based reinforcement learning

The democratization of robotics technology and the development of new actuators progressively bring robots closer to humans. The applications that can now be envisaged drastically contrast with the requirements of industrial robots. In standard manufacturing settings, the criterions used to assess performance are usually related to the robot’s accuracy, repeatability, speed or stiffness. Learning a control policy to actuate such robots is characterized by the search of a single solution for the task, with a representation of the policy consisting of moving the robot through a set of points to follow a trajectory. With new environments such as homes and offices populated with humans, the reproduction performance is portrayed differently. These robots are expected to acquire rich motor skills that can be generalized to new situations, while behaving safely in the vicinity of users. Skills acquisition can no longer be guided by a single form of learning, and must instead combine different approaches to continuously create, adapt and refine policies. The family of search strategies based on expectation-maximization (EM) looks particularly promising to cope with these new requirements. The exploration can be performed directly in the policy parameters space, by refining the policy together with exploration parameters represented in the form of covariances. With this formulation, RL can be extended to a multi-optima search problem in which several policy alternatives can be considered. We present here two applications exploiting EM-based exploration strategies, by considering parameterized policies based on dynamical systems, and by using Gaussian mixture models for the search of multiple policy alternatives.     

Insulation co-ordination for continuous operating voltages and temporary overvoltages

This paper presents an approach to the insulation co-ordination of electric-power installations in the scope of the continuous operating voltages and temporary overvoltages; this approach opens the possibility to calculate the necessary required withstand voltages of unit insulations for short-duration (1 min) power-frequency test voltage, with respect to continuous operating voltages and power-frequency temporary overvoltages and the necessary reliability of the operation of high-voltage power system installations.     

Effective linearization technique for amplifiers operating close to saturation

The linearization technique that uses simultaneous injection of the second harmonics into the amplifier input and their feedforwarding to the amplifier output is suggested. This technique reduces the third‐order intermodulation products in multichannel amplifiers when amplifiers operate closer to saturation. © 2007 Wiley Periodicals, Inc. Int J RF and Microwave CAE, 2007.     

Combining Efficiency and Stability in Mixed Tin–Lead Perovskite Solar Cells by Capping Grains with an Ultrathin 2D Layer

The development of narrow-bandgap (E_g ≈ 1.2 eV) mixed tin–lead (Sn–Pb) halide perovskites enables all-perovskite tandem solar cells. Whereas pure-lead halide perovskite solar cells (PSCs) have advanced simultaneously in efficiency and stability, achieving this crucial combination remains a challenge in Sn–Pb PSCs. Here, Sn–Pb perovskite grains are anchored with ultrathin layered perovskites to overcome the efficiency-stability tradeoff. Defect passivation is achieved both on the perovskite film surface and at grain boundaries, an approach implemented by directly introducing phenethylammonium ligands in the antisolvent. This improves device operational stability and also avoids the excess formation of layered perovskites that would otherwise hinder charge transport. Sn–Pb PSCs with fill factors of 79% and a certified power conversion efficiency (PCE) of 18.95% are reported—among the highest for Sn–Pb PSCs. Using this approach, a 200-fold enhancement in device operating lifetime is achieved relative to the nonpassivated Sn–Pb PSCs under full AM1.5G illumination, and a 200 h diurnal operating time without efficiency drop is achieved under filtered AM1.5G illumination.     

Evaluation of the energy saving opportunities for palm oil refining process: Sahabat Oil Products (SOP) in Lahad Datu,Malaysia

Palm Oil industry is one of the major contributors to Malaysia’s economic activity. Accounting for 39 % of the world palm oil production and 44 % of world exports, Malaysia holds an important niche in fulfiling the growing global needs for oils and fats sustainably. This industry has high potential for further improvements especially in terms of energy saving as a major contributor to cost and emission reduction. An analysis of the refining process of palm oil in Sahabat Oil Products, Lahad Datu has been performed and presented in this paper for scoping potential energy and cost savings using heat integration. A first stage optimisation of the minimum temperature difference, ?T _min, of a heat exchanger network (HEN) has been performed. The goal has been to evaluate the maximal possible heat recovery as well as the appropriate placement of utilities. The HEN design is presented in both grid diagram and shifted retrofit thermodynamic grid diagram (SRTGD). SRTGD representation has been illustrated in this paper as a useful tool for guiding eventual future retrofit. The capital-energy trade-off of the heat recovery targets indicates optimum ?T _min of 12.3 °C. The hot and cold utility targets at ?T _min = 12.3 °C are 1419 and 1649 kW, indicating potential saving of 3.5 and 3.1 % as compared to the existing utility consumption and emissions. Future work could proceed further to seek potentially viable retrofit of the existing heat recovery network.     

Robust nonlinear control of systems with input unmodeled dynamics

We consider systems with unmodeled dynamics appearing at the input, and present a redesign that preserves global asymptotic stability. The admissible unmodeled dynamics are relative degree zero and minimum phase. The redesign is illustrated on an analytical example.     

Influence of an antimicrobial treatment on the strength properties of polyamide/elastane weft‐knitted fabric

Polyamide/elastane weft‐knitted fabric, as a suitable compression material, was treated with gentamicin sulfate and a natural antimicrobial preparation of the autochthonous essential oil of Picea abies. The antimicrobial bioactivity of the treated fabric in vitro was tested for different groups of bacteria and a fungus (Staphylococcus aureus, Escherichia coli, Klabsiella, and Candida albicans). The results of the experiment showed that the antimicrobial‐treated knitted fabric expressed a wide range of bactericidal, fungicidal, and bacteriostatic activity versus the various groups of microorganisms. The influence of the antimicrobial treatments on the tensile strength and elongation, tear strength and elongation, stiffness, and elasticity of the chosen test material was also examined. These properties of the fabric were changed after the antimicrobial treatments. Both the tensile strength and tear strength increased about 14% after the treatment. The scope of the experiment also included the testing of the compression distribution at the skin/knitted fabric interface on the legs of human test subjects. Pressure generated by the application of the knitted fabric on the leg surface increased in correlation with a decrease in the leg circumference. On the basis of these observations, it was concluded that the knitted fabric could be used as a compression textile before and after the treatment. © 2006 Wiley Periodicals, Inc. J Appl Polym Sci 103: 4012–4019, 2007     

A direct method for BBD matrix inversion

Published in Kibernetika i Sistemnyi Analiz, No. 2, pp. 51–64, March–April, 1994.     

Robust output maneuvering for a class of nonlinear systems

The output maneuvering problem involves two tasks. The first, called the geometric task, is to force the system output to converge to a desired path parametrized by a continuous scalar variable θ. The second task, called the dynamic task, is to satisfy a desired dynamic behavior along the path. This dynamic behavior is further specified via a time, speed, or acceleration assignment. While the main concern is to satisfy the geometric task, the dynamic task ensures that the system output follows the path with the desired speed. A robust recursive design technique is developed for uncertain nonlinear plants in vectorial strict feedback form. First the geometric part of the problem is solved. Then an update law is constructed that bridges the geometric design with the speed assignment. The design procedure is illustrated through several examples.     

Cryogels of cellulose derivatives prepared via UV irradiation of moderately frozen systems

Cryogels of non-ionic (hydroxypropyl)methylcellulose, 2-hydroxyethylcellulose, methylcellulose and cationic hydroxyethylcelluloses were prepared for the first time via UV irradiation of moderately frozen systems. The influence of the irradiation time, the concentration of the photoinitiator (4-benzoylbenzyl)trimethylammonium chloride, and the concentration of the polymer solution on the efficiency of the cross-linking process in moderately frozen systems was investigated. It was found that 2-min irradiation at an irradiation dose rate of 5.7 J/cm² min is adequate for cross-linking of 3 wt.% polymer solutions resulting in high quality cryogel.