On-Road Driver Emotion Recognition Using Facial Expression

With the development of intelligent automotive human-machine systems, driver emotion detection and recognition has become an emerging research topic. Facial expression-based emotion recognition approaches have achieved outstanding results on laboratory-controlled data. However, these studies cannot represent the environment of real driving situations. In order to address this, this paper proposes a facial expression-based on-road driver emotion recognition network called FERDERnet. This method divides the on-road driver facial expression recognition task into three modules: a face detection module that detects the driver’s face, an augmentation-based resampling module that performs data augmentation and resampling, and an emotion recognition module that adopts a deep convolutional neural network pre-trained on FER and CK+ datasets and then fine-tuned as a backbone for driver emotion recognition. This method adopts five different backbone networks as well as an ensemble method. Furthermore, to evaluate the proposed method, this paper collected an on-road driver facial expression dataset, which contains various road scenarios and the corresponding driver’s facial expression during the driving task. Experiments were performed on the on-road driver facial expression dataset that this paper collected. Based on efficiency and accuracy, the proposed FERDERnet with Xception backbone was effective in identifying on-road driver facial expressions and obtained superior performance compared to the baseline networks and some state-of-the-art networks.

Ontology-Enriched Specifications Enabling Findable, Accessible, Interoperable, and Reusable Marine Metagenomic Datasets in Cyberinfrastructure Systems

Marine microbial ecology requires the systematic comparison of biogeochemical and sequence data to analyze environmental influences on the distribution and variability of microbial communities. With ever-increasing quantities of metagenomic data, there is a growing need to make datasets Findable, Accessible, Interoperable, and Reusable (FAIR) across diverse ecosystems. FAIR data is essential to developing analytical frameworks that integrate microbiological, genomic, ecological, oceanographic, and computational methods. Although community standards defining the minimal metadata required to accompany sequence data exist, they haven’t been consistently used across projects, precluding interoperability. Moreover, these data are not machine-actionable or discoverable by cyberinfrastructure systems. By making ‘omic and physicochemical datasets FAIR to machine systems, we can enable sequence data discovery and reuse based on machine-readable descriptions of environments or physicochemical gradients. In this work, we developed a novel technical specification for dataset encapsulation for the FAIR reuse of marine metagenomic and physicochemical datasets within cyberinfrastructure systems. This includes using Frictionless Data Packages enriched with terminology from environmental and life-science ontologies to annotate measured variables, their units, and the measurement devices used. This approach was implemented in Planet Microbe, a cyberinfrastructure platform and marine metagenomic web-portal. Here, we discuss the data properties built into the specification to make global ocean datasets FAIR within the Planet Microbe portal. We additionally discuss the selection of, and contributions to marine-science ontologies used within the specification. Finally, we use the system to discover data by which to answer various biological questions about environments, physicochemical gradients, and microbial communities in meta-analyses. This work represents a future direction in marine metagenomic research by proposing a specification for FAIR dataset encapsulation that, if adopted within cyberinfrastructure systems, would automate the discovery, exchange, and re-use of data needed to answer broader reaching questions than originally intended.

Modeling the Interaction of Forest Machines with Soil when Working on Slopes

The urgency of the task of effective development of cutting areas on the slopes of mountains, hills, and hilly-ridge reliefs is primarily due to the depletion of available, special, lowland operational forests in Siberia and the Far East, which were once, not quite correctly, called forest-surplus regions of the Russian Federation. The operational woodlands that are convenient for development in Siberia and the Far East are largely depleted. To develop new ones, large-scale road construction is necessary, which requires significant financial expenses and reduces the already low profitability of logging production. It is also declining due to the ever-increasing volume of export of harvested timber, even if the cost of construction and maintenance of a new network of logging roads is not considered. Forest ecosystems located on slopes are among the most vulnerable. When working on the slopes with traditional systems of logging machines, it becomes necessary to cut a serpentine of skid trails, which later become concentrators of water and wind erosion. Currently, leading manufacturers of machinery and equipment for the forest industry, such as Ponsse, John Deer, Komatsu, and others, have developed technical solutions that significantly facilitate the operation of forest machine systems. Such solutions, first of all, include winches integrated into the transmissions of machines. Another solution is to use separate self-propelled winches remotely controlled by the operator of a forest machine, for example, T-winch. In this case, the machine does not receive additional weight from the winch integrated into it; however, the negative impact of forest machines on soils does not disappear, but has its own significant specifics. The article shows that when performing logging operations on slopes, primarily steep ones with slope angles exceeding 20–25°, it is necessary to make adjustments to the assessment of the destruction nature of the soil array and the patterns of the track depth formation under the influence of the skidding system. For citation: Rudov S.E., Shapiro V.Ya., Grigorev I.V., Kunitskaya O.A., Grigoreva O.I. Modeling the Interaction of Forest Machines with Soil when Working on Slopes. Lesnoy Zhurnal [Russian Forestry Journal], 2021, no. 6, pp. 121–134. DOI: 10.37482/0536-1036-2021-6-121-134

Intelligent control based on ergatic systems in conditions of incomplete and fuzzy information

The article deals with the management of human-machine (ergatic) systems in the conditions of digital transformation in relation to their functioning in the presence of NON-factors, such as: uncertainty, complexity, instability, ambiguity. Modern conditions for the formation of the digital economy imply the search and use of a new methodology in the organization of management activities, including the regional level. This process is carried out through the widespread use of human-machine systems with a high level of intellectualization of the machine component, the use of hybrid intelligence and the formation of bionts. We show a variant of classification of ergatic systems, focused on socio-economic systems. We propose a method for choosing a rational alternative to support the management of human-machine systems in the conditions of vagueness and ambiguity of the initial data and approaches to the management quality criteria. A fuzzy approach to a multi-criteria problem is proposed. It leads to a certain combination of fuzzy selection criteria, and to the study of complex systems as a hierarchical structure, with the representation of system elements and its qualitative properties as fuzzy mathematical models, the combination of which will give a mathematical model of the systems.

DTC with fuzzy logic for multi-machine systems: traction applications

In this work, a direct torque control (DTC) method for multi-machine systems is applied to electric vehicles (EVs). Initially, the DTC control method associated with the model reference adaptive system (MRAS) is used for speed control, and management of the magnetic quantities is ensured by the variable master-slave control (VMSC). In order to increase the technical performance of the studied system, a DTC method has been associated with a fuzzy logic approach. These two control methods are applied to the traction chain of an electric vehicle to highlight its speed, precision, stability, and robustness metric during particular stress tests imposed on the wheel motor. The results obtained in MATLAB/Simulink software made feasible a comparison of two proposed methods based on their technical performances. It should be noted that the direct fuzzy logic torque control (DFTC) has better performance than the DTC associated with the MRAS system as a rise time reduction of 1.4%, an oscillation of torque, and flux amplitude of less than 9%, static steady-state error near zero. The DTFC control method responds favorably to electric vehicle traction chain systems by the nature of the comfort and safety provided.

A brief history of post-war soviet engineering psychology and ergonomics in persons

The article deals with a brief history in the faces of post-war engineering psychology and ergonomics. It is shown that the emergence of disciplines in the Soviet Union taking into account the human factor while building complex systems of guided weapons and their improvement was an organized process of specialists’ joint activities in a wide class of scientific and practical disciplines involved in designing and operating ergatic systems for general and military purposes. Prominent representatives of the engineering-psychological and ergonomic discourse of the Soviet engineering psychology and ergonomics of the post-war period are considered in historical and personal terms. Their contribution to the theory and practice of designing man-machine systems is considered. The article shows the existence of local scientific schools in the field of human factor problems, the schools have arisen around the authoritative leaders and scholars solving scientific, design and organizational tasks for creating a new technology. The main literature sources, which served as a scientific and educational-methodological basis for developing domestic engineering psychology and ergonomics, are highlighted. The centres of advancing academic research and applying scientific developments into practice in the field of engineering psychology and ergonomics in modern Russia are stressed.