This article describes a series of experiments in the Gazebo simulation environment aimed at studying the influence of external weather conditions on the automatic landing of an unmanned aerial vehicle (UAV) on a moving platform using computer vision and a previously developed control system based on PID and polynomial controllers. As part of the research, methods for modeling external weather conditions were developed and landing tests were carried out simulating weather conditions such as wind, light, fog and precipitation, including their combinations. In all experiments, successful landing on the platform was achieved; during the experiments, landing time and its accuracy were measured. The graphical and statistical analysis of the obtained results revealed the influence of illumination, precipitation and wind on the UAV landing time, and the introduction of wind into the simulation under any other external conditions led to the most significant increase in landing time. At the same time, the study failed to identify a systemic negative influence of external conditions on landing accuracy. The results obtained provide valuable information for further improvement of autonomous automatic landing systems for UAVs without the use of satellite navigation systems.

Purpose of research.Emotions play one of the key roles in the regulation of human behaviour. Solving the problem of automatic recognition of emotions makes it possible to increase the effectiveness of operation of a whole range of digital systems such as security systems, human-machine interfaces, e-commerce systems, etc. At the same time, the low efficiency of modern approaches to recognizing emotions in speech can be noted. This work studies automatic recognition of emotions in speech applying machine learning methods.Methods.The article describes and tests an approach to automatic recognition of emotions in speech based on multitask learning of deep convolution neural networks of AlexNet and VGG architectures using automatic selection of the weight coefficients for each task when calculating the final loss value during learning. All the models were trained on a sample of the IEMOCAP dataset with four emotional categories of ‘anger’, ‘happiness’, ‘neutral emotion’, ‘sadness’. The log-mel spectrograms of statements processed by a specialized algorithm are used as input data.Results.The considered models were tested on the basis of numerical metrics: the share of correctly recognized instances, accuracy, completeness, f-measure. For all of the above metrics, an improvement in the quality of emotion recognition by the proposed model was obtained in comparison with the two basic single-task models as well as with known solutions. This result is achieved through the use of automatic weighting of the values of the loss functions from individual tasks when forming the final value of the error in the learning process.Conclusion.The resulting improvement in the quality of emotion recognition in comparison with the known solutions confirms the feasibility of applying multitask learning to increase the accuracy of emotion recognition models. The developed approach makes it possible to achieve a uniform and simultaneous reduction of errors of individual tasks, and is used in the field of emotions recognition in speech for the first time.