7 results about "Software error" patented technology

The application discloses a task-level redundant fault-tolerant system based on an RTOS kernel, which comprises a user API layer, a redundant task manager, a state synchronization engine and a fault detection and switching controller, and a shared memory partition. The user API layer is used for creating a redundant task group by calling a special API interface. The redundant task manager is used for managing the whole life cycle of the redundant task group. The state synchronization engine is used for synchronizing state data and buffering task input data. The fault detection and switching controller is used for detecting timeout, software errors and hardware abnormal faults, and switching the main task and the backup task after detecting the failure. The shared memory partition is used for synchronizing information between the main task and the backup task, and modifying the access authority when the main task and the backup task are switched by configuring the access authority by an MPU. The task-level redundant fault-tolerant system based on the RTOS kernel disclosed by the application reduces the development and maintenance cost, improves the fault identification accuracy, and guarantees the execution continuity after the main-backup switching.

A computer device for predicting software errors based on artificial intelligence, wherein the device comprises the following: at least one processing unit that is operationally coupled with a non-volatile storage unit; a data acquisition unit configured to receive software development artifacts such as source code files, change histories, execution logs, compilation results, and error notes from one or more external computer systems via a communication interface; a preprocessing processor configured to normalize received software development artifacts and structurally transform them into standardized data representations by performing token separation, structural parsing, and metric alignment operations; a feature extraction processor that is operationally connected to the preprocessing processor and is configured to generate multidimensional feature representations that correspond to software complexity attributes, dependency relationships, code change frequency, and execution behavior patterns; a learning processor that is stored in non-volatile memory and executable by the processing unit, wherein the learning processor is configured to create predictive models by iterative parameter adjustment using historical feature representations and corresponding error indicators; a prediction processor configured to apply prediction models to newly generated feature representations to determine a probability value associated with the occurrence of software errors in identified software components; a feedback processing unit configured to receive confirmed error information and update the predictive models by integrating newly acquired error results into stored training data; and an output interface unit configured to output prediction results, indicating the probability of errors, affected software components, and severity estimates, to an external computing environment.

The invention provides a sequential control circuit and method for a main motor and an auxiliary oil pump, and relates to the technical field of industrial electrical control. According to the invention, an expensive programmable logic controller (PLC) or a decentralized control system (DCS) is abandoned, and only conventional industrial electrical elements such as a time relay, an intermediate relay, a contactor and the like are adopted to construct control logic. The circuit structure is clear, elements are universal and easy to obtain, and the hardware purchase cost of equipment and the overall system complexity are remarkably reduced. The control logic is completely realized by a hardware circuit and does not depend on any software program, so that the risk of control logic disorder caused by program loss, software errors or virus infection and the like is fundamentally eradicated. A hardware circuit is strong in execution certainty, good in anti-electromagnetic interference capability, and more stable and reliable in system operation. The method is particularly suitable for single machines such as steam turbines, supergravity machines and emulsifying machines or large-scale equipment which is sensitive to cost and needs strict start-stop time sequence protection.

The invention discloses a software error code statistical method and device and a related medium, and the method comprises the steps: determining a plurality of field types based on error reporting configuration, and obtaining an error code coding rule object; performing error return body field definition according to the error code coding rule object, and configuring an error processing rule corresponding to each module to obtain an error processing rule set; performing link tracking identifier generation and transparent transmission configuration according to the error processing rule set to obtain a link tracking transparent transmission configuration object; performing structured analysis on the error log data corresponding to the link tracking transparent transmission configuration object to obtain an analysis field data set; and performing statistical analysis processing on the analysis field data set to obtain an error code statistical analysis result. According to the method and the device, the obtained error log data is subjected to structured analysis and statistical analysis processing to obtain the error code statistical analysis result, so that the source of the error code can be positioned, and the accuracy of the error code statistical result is improved.

The application provides a safety system and a vehicle, which are applied to the technical field of vehicles and aim to solve the problem of how to improve the safety of a real-time operating system. The safety system comprises a real-time operating system, a functional safety island (FSI) and a controller. The FSI is in communication connection with the real-time operating system and the controller respectively. The FSI is used for monitoring faults of the real-time operating system and uploading fault data to the controller in the case that the real-time operating system has a fault. The fault data comprises at least one of software error data and hardware fault data. The controller is used for performing corresponding safety disposal operations according to the received fault data. The safety system provided by the application can discover faults in the real-time operating system in time through the FSI and improve the safety of the real-time operating system.

This invention relates to the field of software testing technology, and proposes a software bug-planting method and system based on software requirements. The method includes the following steps: analyzing software requirements to determine software requirement characteristics and their characteristic states; determining combinations of characteristic states for multiple software requirement characteristics; identifying valid characteristic state combinations to generate identified characteristic state combinations, and determining the software output of the identified characteristic state combinations; constructing software errors for the identified characteristic state combinations based on the software output; and using the software errors to bug-plant the source program to generate bug-planted software. This invention can assess the software test coverage of software requirements based on software requirements, and simultaneously assess the ability of software testing to discover software problems using the bug-planted source program as the implementation carrier. Therefore, it can simultaneously assess and accept the two core objectives of software testing, greatly improving the effectiveness and sufficiency of software test acceptance.

The invention discloses a software error version positioning method and related equipment, and the method comprises the steps: executing a regression test on all software change versions of target test software, and obtaining original error traceability data corresponding to the target test software when detecting that error software versions greater than a preset error number exist; performing feature vectorization processing on the original error traceability data to obtain a target feature vector; inputting the target feature vector into a large model for error probability analysis, and generating error version probability data; and determining an initial error software version based on the error version probability data, and generating a positioning analysis report corresponding to the initial error software version. According to the method, the positioning speed and accuracy of the error version can be improved, the manual troubleshooting pressure is reduced, and the method can be widely applied to the fields of animation game making engine software, development systems, game animation software and the like in the digital creative industry.