Anoush Sepehri
I am a Mechanical Engineering Ph.D. student at the University of California, San Diego where I am co-supervised by Prof. Michael Tolley and Prof. Tania Morimoto. My research interests are in the design, and control of new wearable soft robotic systems to address challenges in rehabilitation, human assistance, and haptics.
Before enrolling at UC San Diego, I was a Mechanical Engineering student at the University of British Columbia where I worked on a number of projects in surgical robotics, tactile sensing, and 3D bioprinting.
Education
University of California, San Diego
Doctor of Philosophy - Ph.D., Mechanical Engineering
2022 - 2027
Co-supervised by Prof. Michael Tolley and Prof. Tania Morimoto
Supported by the UCSD MAE First-Year Fellowship and the Natural Sciences and Engineering Research Council of Canada Postgraduate Scholarship - Doctoral Program (NSERC PGS-D)
University of British Columbia
Bachelor's of Applied Science - B.A.Sc., Mechanical Engineering
2017 - 2022
Specialization in Mechatronics
Recipient of the Natural Sciences and Engineering Research Council of Canada Undergraduate Student Research Award (NSERC USRA), BioTalent Student Work Placement Award, and Trek Excellence Scholarship.
Research
High Force and High Bandwidth Thermal Actuator Bundles using Liquid Crystal Elastomers Units
Coming Soon!
Retrofitting Soft Assistive Robots with Sew-free Sensing Garments for Joint Motion Tracking and Kinematic Feedback
Coming Soon!
A Soft Robotic Wrist Orthosis using Textile
Pneumatic Actuators for Passive Rehabilitation
Anoush Sepehri, Samual Ward, Michael T. Tolley, Tania K. Morimoto
2024 IEEE 7th International Conference on Soft Robotics (RoboSoft)
(Best Paper Finalist)
In this paper, we developed a wearable robotic orthosis for passive flexion and extension rehabilitation of the wrist. The goal of this work was the capture the functionality of a standard continuous passive motion device for the wrist in a system that is soft, accessible, easy to use, and comfortable for long periods of wear. We developed textile pneumatic actuators that can conform to the wrist joint and demonstrate that our device can achieve a larger range of motion at lower operating pressures compared to existing soft, pneumatically actuated devices.
A Pneumatically Actuated Cutaneous Fingertip Device for Surface Perception
Final course project for UCSD MAE 207: Haptic Interfaces
For this class, we developed a haptic device that can render both normal and shear forces on the fingertip. The device consisted of a series of pneumatic pouches that move a tactor around to generate different levels of skin stretch and normal force on the fingertip. Our device was capable of rendering various surface properties and demonstrated the importance of cutaneous feedback for virtual reality and surface perception.
Anoush Sepehri, Hamed Helisaz, Mu Chiao
Sensors and Actuators A: Physical (2023)
In this work, we presented a tactile sensor that used fiber Bragg grating technology for tissue palpation, specifically for the diagnosis of prostate cancer. We conducted ex-vivo palpation experiments with our sensor and were able to distinguish between healthy and cancerous phantom tissues using a quasilinear viscoelastic material model that analyzed the elastic and viscous properties of soft tissue based on the sensor readings during ramp and hold compressions.
UBC Open Library (2022)
For my undergraduate research thesis, I developed a closed-loop controlled robotic base that can be retrofitted onto existing C-arms to aid in intra-operative fluoroscopic imaging during orthopedic surgeries. I accomplished this by working on two technologies previously developed in Prof. Hodgson's research group, the Easy-C and the OPTIX. I conducted preliminary experiments that demonstrated that we were able to achieve superior positioning accuracy and repeatability in comparison to manual positioning for common movements frequently done during surgeries.
Anoush Sepehri, Amirreza M. Moghaddam
IEEE Access (2021)
In this paper, we presented a motion planning algorithm for redundant manipulators that combined rapidly exploring randomized trees for end-effector path planning with artificial potential fields for joint trajectory planning. We demonstrated the algorithm's efficacy in both simulated and physical experiments and showed that we can avoid local minima during joint trajectory planning when solely relying on artificial potential fields.