Omid MahdiZadeh - Experiences

Omid MahdiZadeh

Experiences

Research Experiences

Sina Robotics and Medical Innovators Co., LTD.

✓Head of surgical robots control Group

✓ Configuration setup for workspace optimization

✓ Force sensing and Robot surgery haptic assistance

✓ Robot motion safety and reliability in surgery operation

✓ Achieved the kinematics and dynamics model SINA Robot(Tele Surgical Robot)

Advanced Robotics & Automated Systems (ARAS)
Center of Excellence in Robotics and Control,
Advanced Robotics & Automated Systems Lab (ARAS)

✓Research Assistant

✓ Design and Manufacturing Measurement
and electronic subsystem of RoboWalk

✓ Appropriate observer and estimation states system

✓ Precise attitude determination and fault detection

✓ Design and manufacturing of a 6 DOF force measuring

✓ Achieved the kinematics and dynamics model
of a spherical parallel manipulator

" OMID ROBOTICS TEAM " (ORT)
Team in Small Size Soccer League

✓Head of AI Group

✓ Manufacturing Hexacopter Drone

✓ Decision-making process for Game Playing

✓ Path planning of the Small Size Soccer Robot

✓ Implement feed-forward control model based

" Unbounded Designers "
Team in Humanoid League

✓ AI member

✓ Localization and mapping

✓ Humanoid Robot Vision processing

✓ Ball, Robot, and distance detection

Projects

Sina Robotic Telesurgery System

Sina Robotics and Medical Innovators is at the forefront of medical robotics, developing innovative solutions to enhance surgical precision, flexibility, and ergonomics. As a Research Assistant and Head of Surgical Robotics Control Group at Sina Robotics, I contributed to the development of the company's groundbreaking Sina Flex robotic surgery system. I played a pivotal role in optimizing workspace configurations and ensuring robot motion safety and reliability. Additionally, I spearheaded the development of a comprehensive kinematics and dynamics model for the SINA Robot, laying the foundation for accurate system simulation and control.

My expertise lies in developing advanced observer and estimation techniques for surgical robot assistance, optimizing workspace determination and fault detection algorithms, designing and manufacturing a high-precision 6-DOF force-measuring sensor, and developing a comprehensive kinematics and dynamics model for the SINA Robot (Tellsurgical robot).

With my contributions to the Sina Flex robotic surgery system, I have demonstrated my expertise in developing innovative control strategies, optimizing workspace configurations, integrating force sensing and haptic assistance, and ensuring robotic motion safety and reliability. It is my passion to utilize my skills in robotics research and development to improve patient outcomes and enhance healthcare services.

Sina Robotics and Medical Innovators is committed to innovation and quality, and my work has played a significant role in advancing the company's mission to provide advanced medical equipment that improves the quality of care for patients worldwide.

Multipurpose Test Stand (RoboWalk Assistive Device)

Due to the high cost of implementing physiotherapy and rehabilitation centers, the rehabilitation of partial disability of the lower limbs is being replaced by robotic devices. Wearable assistive devices such as exoskeletons have been recognized as a practical solution to increase the power and maneuverability of human beings in industries, such also rehabilitation, elderly care, and sports exercises. RoboWalk is a lower limb assistive device that has been designed in ARAS lab. To guarantee its proper performance, this test-stand is designed and implemented with extended toolkit options. To validate the range of various scenarios, the electronics boards and subsystems, the correct interaction between the sensors and the actuators with the main control loop and so on are evaluated. Also, stress analyses in SOLIDWORKS software showed that the design of the proposed device is capable to handle the required tasks.

As for the Test-Stand, a personal computer was chosen as the central processing unit. A C++ core code on the PC receives data from Arduino and after organizing it, sends it to a controller program (here Matlab/Simulink) via UDP protocol. In Simulink, all the received data will be subjected to an exponential moving average filter. For a better estimation of the Test-Stand states, a two-stage Kalman filter is used. This Kalman filter as explained in previous studies uses the Encoders and IMUs data for state estimation. By having the system states, the controller algorithm uses this information to calculate control output. This Simulink program allows us to perform various control strategies on the RoboWalk Test-Stand.

Early development stages require a fast and straightforward way to perform first control designs in real-time. In this way, light position control aimed to Test the behavior of the actuator and sensors is implemented. A position controller can also drive a current mode directly to control position as a cascade controller.
In the outer cascade, the position is controlled. The current is controlled in the inner cascade, which must be faster than the control position. knee torque is found by a combination of knee Load Cell and knee Encoder data. Current Mode Control compares the Knee torque with the desired torque from the position loop and modifies the current delivered to the driver to perform the preferred torque.

Lower limb Walking Assistance Exoskeleton (RoboWalk)

Exoskeletons are used in different applications including the area of physical therapy in order to facilitate the patient’s exercises and as an assisting technology to assist the elderly carry out their ordinary activities.

Before constructing the system, the performance of the device must be analyzed through different simulations. Hence, the dynamics of the human and RoboWalk are then obtained using the Newton- Euler (NE) and the Recursive Newton Euler Algorithm (RNEA). The obtained models are then augmented to the human model to estimate the RoboWalk joint forces and torques, and those of the human model. then, RoboWalk is imported to the human model in Opensim software and the augmented model is obtained by defining some constraints and joint models. Controllers are then designed for the human and RoboWalk in Opensim. In the next step, The NE and RNEA models are then compared to the models obtained from Opensim software using the same gait data. It is shown that the NE and RNEA methods match very closely and both of the models possess the same behavior as the Opensim model.

Humanoid Robot

After investigating different methods and algorithms by Baset Teen-Size, MRL-HSL 2017 and Nimbro 2017, we decided to use an Omni-direction based method that can process with different speeds and has been created and developed by Baset Teen-Size. The major improvement of this module is using a trajectory learning approach that was trained on NAO robot in simulation in this module, hands are used to increase dynamic and speed, and to prevent any decrease in stability.

In addition, we slightly modified this module and added a balance control system through force sensitive resistors to improve the performance of this module. We decided to look for various methods of finding the best response to our robot specifications and types of play. Furthermore, we applied a variant of Markov localization, also commonly known as Monte Carlo Localization (MCL), to represent the belief of the robot regarding its current state. This includes the position of the robot in the field and its relative orientation with respect to the middle point. Moreover, each sample consists of a state vector of the underlying system, which can be regarded as a weighting factor and also the pose and angle of the robot in our case. We decided to use a deep learning approach based on the convolutional layer to determine the critical point of landmarks such as the penalty point, cross lines, corner of the field, and goal.

Prarallel Manipulator RoboTab

Due to the control of parallel robots, calculation of forward kinematics is mostly required. In some studies, obtained equations are solved numerically with massive calculations that are not acceptable for control implementations. In this research, to extract forward kinematics for a 3-RRS robot, a new method is introduced. In this method, the kinematics of passive joints is calculated with two separate equations, which can be solved to decrease the total computation demands.
On the one hand, to Upgrade the system, a new weight compensation mechanism is proposed by a linear pneumatic actuator placed between the ground and the end-effector. Since, the system must work quickly for a high frequency task, Convolutional Neural Networks (CNNs) as image processing algorithm, can provide the desired fast feedback.
On the other hand, In order to study the state accuracy and Fault detection reliability of the parallel robot more efficiently and accurately, the Adaptive Unscented Kalman Filter (AUKF) Combined with the Neural Network is used for accurate calibration and fault detection of the system. The 3-RRS parallel robot as a case study is presented to illustrate approximated states with an accuracy better than 1 degree.

Spherical Parallel Manipulator

Spherical parallel manipulators have been proposed for accurate and fast performance. In this Project, kinematics and dynamics of a spherical three degrees-of freedom parallel manipulator are studied. First, both direct and inverse kinematics are solved by using a new geometric approach and defining a new appropriate set of coordinates. Then, based on the derived kinematics equations and Jacobian matrices of links, according to the Lagrange method, the explicit dynamics formulation of the manipulator is developed. The obtained dynamics has been verified by using MSC.ADAMS and MATLAB Software, simultaneously. This explicit verified model is highly important to design and implement various model based control algorithms.

Small Size Soccer Robot

I have been working in the Omid team since 2013. In my first year, I was active in the AI part of the team, and I was an AI leader in the third generation. After a while, I focused on control and mechanics. Now, I advise as a consultant in this Team.

Page updated

Google Sites

Report abuse