SIRIMALLA MANASA

Okay. First of all, thank you for giving this opportunity to introduce about myself. I'm Manasa Sirimalla. I'm working as MATLAB developer with with 2.9 years of experience, and I have experience in simulation, artificial intelligent, machine learning, and simulation, uh, scripting, uh, and scripting using Python and c program and also experience in 3 d point. I have developed many projects in simulation and MATLAB using Python and c programming. And I, uh, developed projects related to the robotics and related to the electrical vehicles, solar systems, and power quality devices. And I completed BTEC in, uh, 2021 in with the 7.8 CGPA and completed diploma in 2018 with 85 percentage. And I'm schooling, uh, in 2015 with 8.5 CGPA. Coming to my family, I have a family of 4 members. My, uh, father is government reseller. My mother is housewife, and my brother is pursuing, uh, diploma. And coming to my strength, uh, I how capable of, uh, learning quickly and, uh, working hard in difficulty time.

Okay. Uh, there are some, uh, strategies to ensure that Python could, uh, return in to to recognize the object in robotics simulation, uh, by using data, argumentation, and normalization, code space transformation, image processing, uh, use of, uh, synthetic data, robust, uh, robust features extinction, and, uh, transfer learning, data balancing, and evaluation and testing, uh, uh, domination, adoption techniques, and, uh, use of reflectional models. Using this above men, uh, mentioned strategies, we need to ensure Python code written for object recognition in robotic simulation to varying lighting conditions.

The techniques, uh, that used in Python to simulate soft body dynamics for robotic applications are by using physics engineers or, uh, finite element method. Next, uh, mesh based deformation. Uh, mesh spring system and also optimization based methods, uh, neural network approaches, and also by using Blender. So by these techniques, we can, uh, simulate the soft body dynamics.

To filters noise from 3 d depth to data, we can use, uh, medi median filtering and, uh, Gaussian filtering, uh, bilateral filtering, uh, statistical, uh, removal, and the grid downs, uh, sampling, temporary filtering, nonlocal, uh, means filtering, and, uh, and, also, come uh, we can filter the noise by combining all, uh, techniques, all filters. Combination of all this filter, also, we can remove the noise from 3 d depth of data to clarity, uh, clarify the simulations.

We can approach, uh, the method, uh, in step by step, uh, step faster. Uh, image acquisition, Uh, to capture the image, we can use this. Uh, next, image processing. Next, uh, edge detection, uh, next thresholding and counter detection, Next, uh, segmentation. Next, object detection. And post. After all this, we can do post processing. Next object classification, By, uh, doing all, uh, by creating all these objects, we can capture the 2 d camera.

The approach that is used to, uh, collision detections, uh, development of collision detection system system in robotics simulation environment using Python by, uh, choose a simulation, uh, enrollment like MATLAB or other software. And, uh, second, set up the simulation enrollment. Next, uh, define collision detection logic. Next, integrate the collision handling. Run the sim uh, after that, we can run the simulation loop. Next, to enhance with the advanced techniques. By using these steps, we can, uh, develop the collision detection system.

In this, uh, given code, the problem is, uh, pointed at the return. Actually, we need to just write the return instead of return none. So there is no need to return none, uh, none code. So instead of that, just we need to write return.

To optimize the processing time of a computer vision algorithms in Python for real time robotic application, The starter, we used, uh, are object detect, uh, uh, algorithm selection and optimization. Next object detection, reduce image size, grayscale conversion, parallelize, uh, computations, or we can use strategies like code optimization. In code optimization, uh, we use optimize already, optimized libraries, next to vectorization, uh, pre computed constants, or can reduce the, uh, function call calls, sir. And another, uh, when we consider hardware, uh, implementation, so, uh, we need to utilize the hardware actualization and reduce the camera resolution. So by using these strategies, we can, uh, optimize the processing time of computer vision.

And there are some strategies to document a complex Python code base, uh, for robotics simulation is, uh, high level overview. So we can, uh, start with the clear and, uh, clear overview and, uh, include the system diagram. And also, uh, we can module level documentation. In this, we can, uh, document each module, uh, and add comments for each, uh, code. Next to detailed code explanation. So in this detailed code explanation, it includes, uh, clarity of, uh, focus on clarity and, uh, examples, usage, and, uh, read me file. Next, uh, collaboration and communication. So in this, uh, we can control the version, code reviews, communication channels. And, also, we can, uh, take the tips from the other team members or, uh, audience, and we can maintain the, uh, or we can keep the documentation proper. And we can, uh, also use automate documentation generation to ex, uh, to explore more information and to maintain the high efficiency of the system.

The steps that we follow to perform regression testing and newly introduced to computer vision features is, uh, first step is identify the test cases. In, uh, in this, we can identify the, uh, functionality functions to create create the test cases. Uh, the this test cases includes both the positive test cases and negative test cases. Next, define expected results. And the next step is, uh, preparation, uh, of, uh, testing enrollment. And the next step is write the pie we can write the Python test scripts. And this test scripts writing by using, uh, uh, testing frameworks, test functions, test function creation, and organize the, uh, uh, test scripts, uh, what we are going to write. And the next step is run the test and analyze the results. So first, we can write the test, then we can analyze result, and we can, uh, compare the, uh, test results with expected results or actual results. And after after testing and analyzing, and the next step is integration with continuous integration. Like, uh, we are integrated that with the CI. Uh, so it can automatically testing the code. And after testing, it is reported, uh, whether it is correct or there any problems associated with the code. So off by using all these steps, we can, uh, perform the regression testing.