Cross-entropy optimization for scaling factors of a fuzzy controller : a see-and-avoid approach for unmanned aerial systems

The Cross-Entropy (CE) is an efficient method for the estimation of rare-event probabilities and combinatorial optimization. This work presents a novel approach of the CE for optimization of a Soft-Computing controller. A Fuzzy controller was designed to command an unmanned aerial system (UAS) for avoiding collision task. The only sensor used to accomplish this task was a forward camera. The CE is used to reach a near-optimal controller by modifying the scaling factors of the controller inputs. The optimization was realized using the ROS-Gazebo simulation system. In order to evaluate the optimization a big amount of tests were carried out with a real quadcopter.

Evaluation of dynamic voltage and frequency scaling for stream programs

Dynamic Voltage and Frequency Scaling (DVFS) offers a huge potential for designing trade-offs involving energy, power, temperature and performance of computing systems. In this paper, we evaluate three different DVFS schemes - our enhancement of a Petri net performance model based DVFS method for sequential programs to stream programs, a simple profile based Linear Scaling method, and an existing hardware based DVFS method for multithreaded applications - using multithreaded stream applications, in a full system Chip Multiprocessor (CMP) simulator. From our evaluation, we find that the software based methods achieve significant Energy/Throughput2(ET−2) improvements. The hardware based scheme degrades performance heavily and suffers ET−2 loss. Our results indicate that the simple profile based scheme achieves the benefits of the complex Petri net based scheme for stream programs, and present a strong case for the need for independent voltage/frequency control for different cores of CMPs, which is lacking in most of the state-of-the-art CMPs. This is in contrast to the conclusions of a recent evaluation of per-core DVFS schemes for multithreaded applications for CMPs.

Cross-entropy optimization for scaling factors of a fuzzy controller: a see-and-avoid approach for unmanned aerial systems

The Cross-Entropy (CE) is an efficient method for the estimation of rare-event probabilities and combinatorial optimization. This work presents a novel approach of the CE for optimization of a Soft-Computing controller. A Fuzzy controller was designed to command an unmanned aerial system (UAS) for avoiding collision task. The only sensor used to accomplish this task was a forward camera. The CE is used to reach a near-optimal controller by modifying the scaling factors of the controller inputs. The optimization was realized using the ROS-Gazebo simulation system. In order to evaluate the optimization a big amount of tests were carried out with a real quadcopter.

UAS See-and-Avoid Strategy using a Fuzzy Logic Controller Optimized by Cross-Entropy in Scaling Factors and Membership Functions

This work aims to develop a novel Cross-Entropy (CE) optimization-based fuzzy controller for Unmanned Aerial Monocular Vision-IMU System (UAMVIS) to solve the seeand- avoid problem using its accurate autonomous localization information. The function of this fuzzy controller is regulating the heading of this system to avoid the obstacle, e.g. wall. In the Matlab Simulink-based training stages, the Scaling Factor (SF) is adjusted according to the specified task firstly, and then the Membership Function (MF) is tuned based on the optimized Scaling Factor to further improve the collison avoidance performance. After obtained the optimal SF and MF, 64% of rules has been reduced (from 125 rules to 45 rules), and a large number of real flight tests with a quadcopter have been done. The experimental results show that this approach precisely navigates the system to avoid the obstacle. To our best knowledge, this is the first work to present the optimized fuzzy controller for UAMVIS using Cross-Entropy method in Scaling Factors and Membership Functions optimization.

UAS See-and-Avoid Strategy using a Fuzzy Logic Controller Optimized by Cross-Entropy in Scaling Factors and Membership Functions.

This work aims to develop a novel Cross-Entropy (CE) optimization-based fuzzy controller for Unmanned Aerial Monocular Vision-IMU System (UAMVIS) to solve the seeand-avoid problem using its accurate autonomous localization information. The function of this fuzzy controller is regulating the heading of this system to avoid the obstacle, e.g. wall. In the Matlab Simulink-based training stages, the Scaling Factor (SF) is adjusted according to the specified task firstly, and then the Membership Function (MF) is tuned based on the optimized Scaling Factor to further improve the collison avoidance performance. After obtained the optimal SF and MF, 64% of rules has been reduced (from 125 rules to 45 rules), and a large number of real flight tests with a quadcopter have been done. The experimental results show that this approach precisely navigates the system to avoid the obstacle. To our best knowledge, this is the first work to present the optimized fuzzy controller for UAMVIS using Cross-Entropy method in Scaling Factors and Membership Functions optimization.

The Role of Anharmonicity in Hydrogen-Bonded Systems: The Case of Water Clusters

The nature of vibrational anharmonicity has been examined for the case of small water clusters using second-order vibrational perturbation theory (VPT2) applied on second-order Møller–Plesset perturbation theory (MP2) potential energy surfaces. Using a training set of 16 water clusters (H2O)n=2–6,8,9 with a total of 723 vibrational modes, we determined scaling factors that map the harmonic frequencies onto anharmonic ones. The intermolecular modes were found to be substantially more anharmonic than intramolecular bending and stretching modes. Due to the varying levels of anharmonicity of the intermolecular and intramolecular modes, different frequency scaling factors for each region were necessary to achieve the highest accuracy. Furthermore, new scaling factors for zero-point vibrational energies (ZPVE) and vibrational corrections to the enthalpy (ΔHvib) and the entropy (Svib) have been determined. All the scaling factors reported in this study are different from previous works in that they are intended for hydrogen-bonded systems, while others were built using experimental frequencies of covalently bonded systems. An application of our scaling factors to the vibrational frequencies of water dimer and thermodynamic functions of 11 larger water clusters highlights the importance of anharmonic effects in hydrogen-bonded systems.

Benchmark Structures and Binding Energies of Small Water Clusters with Anharmonicity Corrections

For (H2O)n where n = 1–10, we used a scheme combining molecular dynamics sampling with high level ab initio calculations to locate the global and many low lying local minima for each cluster. For each isomer, we extrapolated the RI-MP2 energies to their complete basis set limit, included a CCSD(T) correction using a smaller basis set and added finite temperature corrections within the rigid-rotor-harmonic-oscillator (RRHO) model using scaled and unscaled harmonic vibrational frequencies. The vibrational scaling factors were determined specifically for water clusters by comparing harmonic frequencies with VPT2 fundamental frequencies. We find the CCSD(T) correction to the RI-MP2 binding energy to be small (<1%) but still important in determining accurate conformational energies. Anharmonic corrections are found to be non-negligble; they do not alter the energetic ordering of isomers, but they do lower the free energies of formation of the water clusters by as much as 4 kcal/mol at 298.15 K.

The Role of Anharmonicity in Hydrogen-Bonded Systems: The Case of Water Clusters

The nature of vibrational anharmonicity has been examined for the case of small water clusters using second-order vibrational perturbation theory (VPT2) applied on second-order Møller–Plesset perturbation theory (MP2) potential energy surfaces. Using a training set of 16 water clusters (H2O)n=2–6,8,9 with a total of 723 vibrational modes, we determined scaling factors that map the harmonic frequencies onto anharmonic ones. The intermolecular modes were found to be substantially more anharmonic than intramolecular bending and stretching modes. Due to the varying levels of anharmonicity of the intermolecular and intramolecular modes, different frequency scaling factors for each region were necessary to achieve the highest accuracy. Furthermore, new scaling factors for zero-point vibrational energies (ZPVE) and vibrational corrections to the enthalpy (ΔHvib) and the entropy (Svib) have been determined. All the scaling factors reported in this study are different from previous works in that they are intended for hydrogen-bonded systems, while others were built using experimental frequencies of covalently bonded systems. An application of our scaling factors to the vibrational frequencies of water dimer and thermodynamic functions of 11 larger water clusters highlights the importance of anharmonic effects in hydrogen-bonded systems.

Compiler-Directed Frequency and Voltage Scaling for a Multiple Clock Domain

Multiple Clock Domain processors provide an attractive solution to the increasingly challenging problems of clock distribution and power dissipation. They allow their chips to be partitioned into different clock domains, and each domain’s frequency (voltage) to be independently configured. This flexibility adds new dimensions to the Dynamic Voltage and Frequency Scaling problem, while providing better scope for saving energy and meeting performance demands. In this paper, we propose a compiler directed approach for MCD-DVFS. We build a formal petri net based program performance model, parameterized by settings of microarchitectural components and resource configurations, and integrate it with our compiler passes for frequency selection.Our model estimates the performance impact of a frequency setting, unlike the existing best techniques which rely on weaker indicators of domain performance such as queue occupancies(used by online methods) and slack manifestation for a particular frequency setting (software based methods).We evaluate our method with subsets of SPECFP2000,Mediabench and Mibench benchmarks. Our mean energy savings is 60.39% (versus 33.91% of the best software technique)in a memory constrained system for cache miss dominated benchmarks, and we meet the performance demands.Our ED2 improves by 22.11% (versus 18.34%) for other benchmarks. For a CPU with restricted frequency settings, our energy consumption is within 4.69% of the optimal.

Compiler-Directed Dynamic Voltage Scaling using Program Phases

Energy consumption has become a major constraint in providing increased functionality for devices with small form factors. Dynamic voltage and frequency scaling has been identified as an effective approach for reducing the energy consumption of embedded systems. Earlier works on dynamic voltage scaling focused mainly on performing voltage scaling when the CPU is waiting for memory subsystem or concentrated chiefly on loop nests and/or subroutine calls having sufficient number of dynamic instructions. This paper concentrates on coarser program regions and for the first time uses program phase behavior for performing dynamic voltage scaling. Program phases are annotated at compile time with mode switch instructions. Further, we relate the Dynamic Voltage Scaling Problem to the Multiple Choice Knapsack Problem, and use well known heuristics to solve it efficiently. Also, we develop a simple integer linear program formulation for this problem. Experimental evaluation on a set of media applications reveal that our heuristic method obtains a 38% reduction in energy consumption on an average, with a performance degradation of 1% and upto 45% reduction in energy with a performance degradation of 5%. Further, the energy consumed by the heuristic solution is within 1% of the optimal solution obtained from the ILP approach.

Regional processes in mangrove ecosystems: spatial scaling relationships, biomass, and turnover rates following catastrophic disturbance

Physiological processes and local-scale structural dynamics of mangroves are relatively well studied. Regional-scale processes, however, are not as well understood. Here we provide long-term data on trends in structure and forest turnover at a large scale, following hurricane damage in mangrove ecosystems of South Florida, U.S.A. Twelve mangrove vegetation plots were monitored at periodic intervals, between October 1992 and March 2005. Mangrove forests of this region are defined by a −1.5 scaling relationship between mean stem diameter and stem density, mirroring self-thinning theory for mono-specific stands. This relationship is reflected in tree size frequency scaling exponents which, through time, have exhibited trends toward a community average that is indicative of full spatial resource utilization. These trends, together with an asymptotic standing biomass accumulation, indicate that coastal mangrove ecosystems do adhere to size-structured organizing principles as described for upland tree communities. Regenerative dynamics are different between areas inside and outside of the primary wind-path of Hurricane Andrew which occurred in 1992. Forest dynamic turnover rates, however, are steady through time. This suggests that ecological, more-so than structural factors, control forest productivity. In agreement, the relative mean rate of biomass growth exhibits an inverse relationship with the seasonal range of porewater salinities. The ecosystem average in forest scaling relationships may provide a useful investigative tool of mangrove community biomass relationships, as well as offer a robust indicator of general ecosystem health for use in mangrove forest ecosystem management and restoration.

A scaling method for modelling the crashworthiness of novel roadside barrier designs

In this paper, a novel method for modelling a scaled vehicle–barrier crash test similar to the 20◦ angled barrier test specified in EN 1317 is reported. The intended application is for proof-of-concept evaluation of novel roadside barrier designs, and as a cost-effective precursor to full-scale testing or detailed computational modelling. The method is based on the combination of the conservation of energy law and the equation of motion of a spring mass system representing the impact, and shows, for the first time, the feasibility of applying classical scaling theories to evaluation of roadside barrier design. The scaling method is used to set the initial velocity of the vehicle in the scaled test and to provide scaling factors to convert the measured vehicle accelerations in the scaled test to predicted full-scale accelerations. These values can then be used to calculate the Acceleration Severity Index score of the barrier for a full-scale test. The theoretical validity of the method is demonstrated by comparison to numerical simulations of scaled and full-scale angled barrier impacts using multibody analysis implemented in the crash simulation software MADYMO. Results show a maximum error of 0.3% ascribable to the scaling method.

A comparative study on the modified Max-Log-MAP turbodecoding by extrinsic information scaling

A simple but effective technique to improve the performance of the Max-Log-MAP algorithm is to scale the extrinsic information exchanged between two MAP decoders. A comprehensive analysis of the selection of the scaling factors according to channel conditions and decoding iterations is presented in this paper. Choosing a constant scaling factor for all SNRs and iterations is compared with the best scaling factor selection for changing channel conditions and decoding iterations. It is observed that a constant scaling factor for all channel conditions and decoding iterations is the best solution and provides a 0.2-0.4 dB gain over the standard Max- Log-MAP algorithm. Therefore, a constant scaling factor should be chosen for the best compromise.

Factors associated with red blood cell transfusions in very-low-birth-weight preterm infants in Brazilian neonatal units

Preterm infants in neonatal intensive care units frequently receive red blood cells (RBC) transfusions due to the anemia of prematurity. A number of variables related to gestational age, severity of illness and transfusion practices adopted in the neonatal unit where the neonate was born may contribute to the prescription of RBC transfusions. This study aimed to analyse the frequency and factors associated with RBC transfusions in very-low-birth-weight preterm infants. A prospective cohort of 4283 preterm infants (gestational age: 29.9 ± 2.9 weeks; birth weight: 1084 ± 275 g) carried out at 16 university hospitals in Brazil between January 2009 and December 2011 was analysed. Factors associated with RBC transfusions were evaluated using univariate and multiple logistic regression analysis. A total of 2208 (51.6%) infants received RBC transfusions (variation per neonatal unit: 34.1% to 66.4%). RBC transfusions were significantly associated with gestational age (OR: -1.098; 95%CI: -1.12 to -1.04), SNAPPE II score (1.01; 1.00-1.02), apnea (1.69; 1.34-2.14), pulmonary hemorrhage (2.65; 1.74-4.031), need for oxygen at 28 days of life (1.56; 1.17-2.08), clinical sepsis (3.22; 2.55-4.05), necrotising enterocolitis (3.80; 2.26-6.41), grades III/IV intraventricular hemorrhage (1.64; 1.05-2.58), mechanical ventilation (2.27; 1.74-2.97), use of umbilical catheter (1.86; 1.35-2.57), parenteral nutrition (2.06; 1.27-3.33), >60 days of hospitalization (5.29; 4.02-6.95) and the neonatal unit where the neonate was born. The frequency of RBC transfusions varied among neonatal intensive care units. Even after adjusting for adverse health conditions and therapeutic interventions, the neonatal unit continued to influence transfusion practices in very-low birth-weight infants.