The paper describes the architecture of software tools for automating the management of code review of software prototypes of design solutions that allow to obtain such effects as ensuring interactive interaction between the designer and an expert performing code review, as well as reducing the expert’s time spent on commenting the code by selecting a comment from the list prepared in advance for each requirement from the requirements obtained during the analysis. The tools being developed consist of three main parts. The first part presents requirements in a question-and-answer format including standard comments to be inserted into the prototype code in case of non-compliance with these requirements. The second one is a relational database, which is designed to store the source codes of prototypes aimed at inspection and passed it with expert comments. The third one is a Web-application that allows the designer to send prototype codes for review and see the results of the inspection, and an expert to provide viewing and editing of the prototype source code with the insertion of both standard, pre-prepared comments, and written in free form.

The article describes an approach to the creation of an experience base for a software design organization, which is focused on its application in the development of software-intensive automated systems (AS). The use of the proposed experience base extends the potential of the OwnWIQA tool-modeling environment in the workplace of a member of the software design team. The specificity of the toolkit includes the design of a reusable model (precedent model) in the process of work execution on the design of an AS, which is included in the experience base for the purpose of its further use in the development of AS or the creation of a new AS from the same family. The article describes the knowledge base model of software projects, tools for finding use cases, implemented in the OwnWIQA environment. Experimental studies are described that reveal the parameters of precision and recall for a precedent search in the knowledge base. The factors influencing these indicators for different users of the system are given. The article may be of interest to specialists in the field of building knowledge and experience bases.

The article describes an approach to the implementation of a software package for determining the probability of a collision of a mobile robot with obstacles. This approach is based on a neural network model with attention. Key feature is the method of the training dataset generation: the labeling of obstacles and the values of the probability of collision with them is performed not in manual mode, but using a deterministic algorithm that uses the result of semantic segmentation using another pre-trained neural network. This method allows to use a poorly detailed description of the external environment for training convolutional neural networks with attention on the example of recognizing obstacles when a mobile robot moves in simulation mode. At the same time, low detail allows to reduce the time-consuming process of manual data labeling due to automatically generated sampling in the NVIDIA Isaac environment, and the attention mechanism allows to increase the interpretability of the analysis results.