Tools

Tools are developing under ASRLAB github page. 

https://github.com/ESOGU-SRLab

- Software Fault Injection Tool - IM-FIT

IM-FIT provides to find the weaknesses on Python and ROS. The user can use IM-FIT with workload and/or code snippets. At the same time the user can create custom workload and code snippets for its codes. The codes scan by IM-FIT to detect the lines. The user can select the lines to use for execution. At the execution modul, the user can select what it wants features to run. The user can show informations about the its tested codes. If the user wants to watch the created scenarios by IM-FIT, it can do it on Gazebo.

http://github.com/ESOGU-SRLAB/imfit

- Camera Fault Injection Tool - CamFITool

This tool is a simple interface that allows injection of image faults into robot cameras. Thanks to this interface, you can create new image libraries by injecting the fault types you have determined, both real-time to TOF and RGB type ROS cameras, and to the image libraries previously recorded by these cameras. For more information about the purpose of this tool: https://arxiv.org/abs/2108.13803

http://github.com/ESOGU-SRLAB/camfitool

- Simulation Based Robot Verification Tool - SRVT 

SRVT can be thought of as a toolkit or advanced method that allows a robotic system to be imported into a simulation environment and applied to validation tests. The basis of the system is the coordinated use of some critical software for the ROS ecosystem. Simulation environment using Gazebo, trajectory planning using Moveit, mission communication and dynamic verification system using ROS Smach package were built in a single ROS package. For more information about the purpose of this tool: https://dergipark.org.tr/en/pub/jster/issue/61588/979689

https://github.com/ESOGU-SRLAB/srvt-ros

CleanAI: DNN Model Quality Evaluation Tool

CleanAI offers a tool for AI developers to optimize the models they are developing. This tool allows the detection of neurons on the model layers that do not affect the decision-making process and the accuracy of the model/have a low impact, and pruning of these neurons from the model. In this way, the aim is to reduce the model size and obtain the same or approximate accuracy values ​​with a lower model size. It provides AI developers with a comprehensive metric report about their models with various neuron coverage metrics.

https://github.com/ESOGU-SRLAB/CleanAI