Research
Designing novel variable impedance actuators (VIA) with adaptive controllers
Our research introduces a novel spiral spring-based SPIRO mechanism with variable impedance capabilities in a compact form. This structure features a double-sided spiral spring for variable stiffness and an innovative air-actuated variable damping system distributed around it. We propose an energy-based sliding-mode controller for precise and safer responses integrating variable damping and energy efficiency via variable stiffness in all compliant actuators. This integrated approach promises versatile solutions for diverse applications requiring adaptable compliance and robust control.
References
[1] S.A Tafrishi, Y. Hirata "SPIRO: A Compliant Spiral Spring–Damper Joint Actuator With Energy-Based Sliding-Mode Controller", IEEE/ASME Transactions on Mechatronics, 2024 Jan 9. (In-press)
[2] S. A. Tafrishi and Y. Hirata, ”A multi-function rolling joint with variable impedance actuator," in Japan Intellectual Property, no. P20210225WO, JP & WIPO, 2022. (pending)
[3] S.A. Tafrishi, X. Dai, Y. Hirata and A. Burns “Discretization and Stabilization of Energy-based Controller for Period Switching Control and Flexible Scheduling”, IEEE American Control Conference (ACC), Atlanta, USA, pp. 844-849, 2022
[4] U. K. Mavinkurve, S.A. Tafrishi, A. Kanada, K. Honda, Y. Nakashima and M. Yamamoto, “An experimental study on energy-based control of rigid parallel series elastic actuator”, IEEE World Automation Congress (WAC), pp. 26-31, 2021.
[5] U. K. Mavinkurve, S.A. Tafrishi, A. Kanada, and M. Yamamoto, “An energy-based control for vibration suppression using an elastic actuator”, IEEE/SICE International Symposium on System Integration (SII), Fukushima, Japan, pp. 600-605, 2021.
[6] S. A. Tafrishi, Experiment video : https://www.youtube.com/watch?v=uOd4rTBAF_I
The CARE framework blockdiagram.
A new "CARE" framework on welfare with cooperative robots under safety and shared autonomy
The framework was developed to bring the healthcare 4.0 concept closer to reality. We presented cooperative passive and active assistive robots with IoT communication protocols. Also, we redesign robot systems and develop a semi-autonomous platform that can perform tasks based on user/patient interaction in real-world care facility scenarios. Our framework provided human–robot interaction under shared autonomy between the user and assisting robots to improve the efficacy of the users in everyday tasks.
References
[1] A. A. Ravankar*, S. A. Tafrishi*, J. Salazar*, F. Seto and Y. Hirata “CARE: Cooperation of AI-Robot Enablers to Create a Vibrant Society”, IEEE Robotics and Automation Magazine (RAM) with ICRA2023 presentation, IEEE, 2022
[2] S.A. Tafrishi, A. A. Ravankar and Y. Hirata “PSM: A Predictive Safety Model for Body Motion Based On the Spring-Damper Pendulum”, IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Kyoto, Japan, 2022.
[3] S.A. Tafrishi, et al. “A Novel Assistive Controller Based on Differential Geometry for Users of the Differential-Drive Wheeled Mobile Robots”, International Conference on Robotics and Automation (ICRA), Philadelphia (PA), USA, 2022.[3] S. A. Tafrishi, Experiment video for ICRA 2022: https://www.youtube.com/watch?v=3iTsNa2hZ9s&t=1s
How to quantify the safety of human motion using a single IMU
In this study, we have proposed a Predictive Safety Model (PSM) to quantify the human's upper-body motion using a single wireless inertial measurement unit. The PSM encompasses a 3-Dofs spring-damper pendulum model that predicts human motion based on a reduced-dimension safety dataset. The estimated safe orientation of humans is obtained by integrating a safety dataset and an elastic spring-damper model so that the proposed approach can realize complex motions at different safety levels.
References
S.A. Tafrishi, A. A. Ravankar and Y. Hirata “PSM: A Predictive Safety Model for Body Motion Based On the Spring-Damper Pendulum”, IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Kyoto, Japan, 2022.
An assistive geometric controller without any desired configuration for operating with mobile robots
This study develops a new assistive geometric controller that improves the user's joystick inputs for mobile robots considering robot velocity and path curvature. The aim was that the assistive controller tries to correct the user inputs under safety constraints by only relying on the IMU sensor and robot sensors. Because the users can not always know the exact goal/configuration, and these goals might change spontaneously, the assistive controller is designed without any requirement of priori desired configuration.
References
[1] S.A. Tafrishi, et al. “A Geometric Assistive Controller for the Users of Wheeled Mobile Robots without Desired States”, IEEE International Conference on Robotics and Biomimetics (ROBIO), pp. 1953-1959, 2021.
[2] S.A. Tafrishi, et al. “A Novel Assistive Controller Based on Differential Geometry for Users of the Differential-Drive Wheeled Mobile Robots”, International Conference on Robotics and Automation (ICRA), Philadelphia (PA), USA, 2022.[3] S. A. Tafrishi, Experiment video for ICRA 2022: https://www.youtube.com/watch?v=3iTsNa2hZ9s&t=1s
Can attaching ferromagnetic materials improve the motion estimation of IMU sensors?
This work presents a novel approach to improving the motion estimation of inertial measurement sensors (IMU) using ferromagnetic materials. We developed a new Kalman filter to fuse the captured data with considering attached ferromagnetic materials to mitigate the external disturbances. The study does a comprehensive evaluation with conventional filters (e.g., AHRS and Rotenberg).
References
[1] S. A. Tafrishi, M. Svinin and M. Yamamoto, "A Motion Estimation Filter for Inertial Measurement Unit With On-Board Ferromagnetic Materials," in IEEE Robotics and Automation Letters with ICRA Presentation, vol. 6, no. 3, pp. 4939-4946, July 2021, doi: 10.1109/LRA.2021.3067301.
[2] S. A. Tafrishi, ICRA Presentation: https://www.youtube.com/watch?v=OaZKEjec8UU&t=3s
Path planning of spin-rolling sphere on different manifolds
The goal of this research is to find a path planning approach that converges rotating convex objects to desired states on different surfaces. We use the Riemannian manifolds and differential geometry (the Darboux frame equations) to develop this new arc-length-based control strategy for the considered path planning problem.
References
[1] S. A. Tafrishi, M. Svinin, M. Yamamoto, & Y. Hirata, "Path Planning of a Spin-Rolling Sphere on a Plane", 2021, arXiv preprint arXiv:2110.12397.
[2] S. A. Tafrishi, M. Svinin, M. Yamamoto, "Darboux-frame-based parametrization for a spin-rolling sphere on a plane: A nonlinear transformation of underactuated system to fully-actuated model." Mechanism and Machine Theory, Elsevier, no. 164, p. 104415, 2021.
[3] S. A. Tafrishi, YouTube: https://www.youtube.com/watch?v=7JK3VtT0aqs
Geometric filter for object detection and tracking
This study has proposed a multilayered geometric filter for edge-based detection. The method uses only a camera and an IMU sensor. The code has been made open-source. The filter has interesting advantages, such as its ability to reduce computation complexity and memory usage when the camera captures large numbers of features in the environment.
References
[1] S. A. Tafrishi and V. E. Kandjani, "Line-Circle: A Geometric Filter for Single Camera Edge-Based Object Detection," 2017 5th RSI International Conference on Robotics and Mechatronics (ICRoM), 2017, pp. 588-594, doi: 10.1109/ICRoM.2017.8466193
[2] S. A. Tafrishi, D. Xiaotian, and V. E. Kandjani, "Line–Circle–Square (LCS): A multilayered geometric filter for edge-based detection." Robotics and Autonomous Systems , no. 137, 2021, p 103732, doi: doi.org/10.1016/j.robot.2021.103732
[3] Youtube Video: https://www.youtube.com/watch?v=hh6B4aF7jDI
Singular-free inverse dynamics for underactuated robots
In this work, we study how we can avoid the inertial-coupling singularities that appear in the inverse dynamics of the underactuated systems. We solve the singular configuration by adding the small-amplitude combined sine waves in the trajectory of the active joint/actuator. We tested our novel approach in two challenging systems at their singular configurations as a) Rolling robot with actuating mass-point, b) 4-DoFs underactuated manipulator with two passive and two active joints.
References:
[1] S. A. Tafrishi, M. Svinin and M. Yamamoto, "Singularity-Free Inverse Dynamics for Underactuated Systems with a Rotating Mass," 2020 IEEE International Conference on Robotics and Automation (ICRA), pp. 3981-3987, 2020 doi: 10.1109/ICRA40945.2020.9197306.
[2] S. A. Tafrishi, M. Svinin and M. Yamamoto, "Inverse dynamics of underactuated planar manipulators without inertial coupling singularities." Multibody System Dynamics", no. 50, pp. 1-23, 2021
(a)
(b)
A novel underactuated rolling robot with fluid propulsion
This work proposes a novel fluid actuated mechanism for rolling robots. This underactuated rolling robot is propelled by rotating spherical masses (cores). In order to displace the spherical masses (cores) in the circular pipes, a fluid circulating internal driving unit is designed.
References
[1] S. A. Tafrishi , E. Esmaeilzadeh, M. Svinin, and M. Yamamoto, "A fluid-actuated driving mechanism for rolling robots." In 2019 IEEE 4th International Conference on Advanced Robotics and Mechatronics (ICARM), pp. 256-261. IEEE, 2019.
[2] S. A. Tafrishi , E. Esmaeilzadeh, M. Svinin, and M. Yamamoto, "Design, modeling, and motion analysis of a novel fluid actuated spherical rolling robot." Journal of Mechanisms and Robotics 11, no. 4, 2019.
[3] S. A. Tafrishi , Y. Bai, E. Esmaeilzadeh, M. Svinin, and M. Yamamoto, "Inverse dynamics-based motion control of a fluid-actuated rolling robot." Russian Journal of Nonlinear Dynamics 15, no. 4, pp. 611-622 2019.
New researches will be updated soon!!!