https://www.saadyousaf.com/about daily 0.75 2025-06-25 https://www.saadyousaf.com/home daily 1.0 2025-08-14 https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/1622956041202-407KC4O7CDGZENU1EQGE/IMG-0152-Facetune-05-06-2021-23-51-25.jpg Home https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/7283648a-c057-426e-843f-775841e0c98a/reneu-robotics-lab_33.jpg Home https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/94122420-c3e7-4b3a-98dc-b62ed84b4ed5/00_kinematics_overview.jpg Home https://www.saadyousaf.com/publications daily 0.75 2025-09-25 https://www.saadyousaf.com/portfolio daily 0.75 2025-06-20 https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/69552eb4-eb59-4d72-9eed-50af8da806d7/00_Soft+Tissue+Indentation+Overview.jpg Portfolio https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/1617767323217-XMWYAA11DI6ETY2OJ3M4/Indenter+Overview.PNG Portfolio https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/afcdb487-1643-4508-8759-3eac3ff5bd8a/00_maestro_overview.jpg Portfolio https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/9cee7a51-b28e-4350-bb58-b0424c8afdd0/00_SEA_schematic_simplified.jpg Portfolio https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/44d96e75-96ce-4030-b09e-63064309dc35/00_strain_gauge_overview.PNG Portfolio https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/b3fe270d-c89b-42dd-a97f-bc7ba3b07870/00_overview_unity_visualization.PNG Portfolio https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/99ec1ecb-0211-4d9e-ac82-f6edc099e05a/00_kinematics_overview.PNG Portfolio https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/b6910945-e047-436c-b4c8-ef7ed6aaab66/Instrumented+Hand+Overview.PNG Portfolio https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/eb8d5e29-5d7e-4e0e-aa36-ca08322da61f/00_actuated_cuff_overview.jpg Portfolio https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/380ee404-c1f3-4fd7-add4-ed76d4fb6731/00_cuff_overview.jpg Portfolio https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/55d514ef-7642-456c-9104-3f34039fd777/00_hand_pHRI_overview.png Portfolio https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/c35b65dd-52e8-4dad-a2c0-766daeaa61d7/00_arm_pHRI_overview.jpg Portfolio https://www.saadyousaf.com/automated-biomechanics-testbed daily 0.75 2025-09-24 https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/136e0d6a-9cdc-4bf0-9150-e7dd29c195cd/Indenter+Overview.PNG Automated Biomechanics Testbed Indenter assembly with the linear indentation carriage mounted on the rotating ring structure, which can be moved at the base. The human limb is placed through the rotating ring and grounded during characterization measurements. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/43dfc046-6397-4118-bbd4-e4647a2cfbc8/Indenter_01.PNG Automated Biomechanics Testbed The linear indentation carriage module includes: (A) linear carriage driven by a stepper motor, (B) single-axis load cell, and (C) indenter head that interfaces with soft tissue. The rotating ring structure includes (D) outer ring, (E) grounded center ring, and (F) inner ring. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/e6bd21ed-7c93-40d6-961f-16291ed02754/Indenter_02.PNG Automated Biomechanics Testbed The experimental setup shows the indentation location fixed at a point along the length of the arm, as well as anchoring cuffs at the wrist and elbow. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/99aee766-73c2-4213-9315-75f318079d80/Indenter_03.PNG Automated Biomechanics Testbed The 12 angular locations measured in this paper (repeated across 3 rows) are shown in the forearm cross section of the right arm. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/4cd0d22f-d413-46e9-b31e-9ff27df6821b/Indenter_04.PNG Automated Biomechanics Testbed Force-displacement response from three trials of indentation at the 30 degree angular location on the first row. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/41266ebb-098b-4e85-9b61-851e2f87fca1/Indenter_05.PNG Automated Biomechanics Testbed Force-displacement response at 6 angular locations on the first row of the forearm for a single human subject. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/a31c96d7-2a10-4f3f-b9ab-4cc56024c9e3/Indenter_06.PNG Automated Biomechanics Testbed Force-displacement response with the forearm relaxed and flexed at two angular locations. Higher stiffness is observed at all angular locations during the flexed case when muscles are co-contracted. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/407db72f-db28-43e3-bcc3-7c00327a560a/Indenter_07.PNG Automated Biomechanics Testbed Viscoelastic relaxation response at two angles. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/aefd47f4-44be-459f-9987-43445fd6c48b/Indenter_08.PNG Automated Biomechanics Testbed Indentation force in response to a sinusoidal input displacement at the 180◦ angular location. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/5fabfba6-8e1c-417a-8fc8-b06f7de78363/Indenter_09.PNG Automated Biomechanics Testbed Human-centered approach for interface design in which interface padding properties are varied depending on stiffness measurements of the human limb. https://www.saadyousaf.com/upper-arm-cuff daily 0.75 2025-09-24 https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/334e932c-3e12-407d-8875-c291c5836ced/Cuff+in+Harmony.PNG Sensorized Arm Cuff The design introduced in this work is implemented for an upper-arm interface in the Harmony exoskeleton. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/ded42f59-477e-48fd-bf2b-6368b68e3b03/Cuff+in+Harmony+2.PNG Sensorized Arm Cuff The sensorized upper-arm cuff is used with the upper-body Harmony exoskeleton and a human user to explore potential applications of this work. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/5fcdbada-d201-496d-a252-101ef91fb9e5/Cuff+Build+Up.PNG Sensorized Arm Cuff The upper-arm cuff uses 24 panels, each covered with foam padding, to allow for comfortable interaction between the wearer and the interface. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/2e63e1e8-7db2-4a27-b740-e7d9cb329916/Cuff+FSR+Locations.PNG Sensorized Arm Cuff Sensor placement on the cuff is distributed across three rows and eight columns. Foam padding is not shown in these images. The orientation of the six-axis load cell can be seen. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/a597488a-f4ff-4b88-9bb8-c674a5f69a0d/Panel+Design.PNG Sensorized Arm Cuff Integration of a single panel (white) installed in the sensorized cuff (blue) shows how it interfaces with the FSR (green) through a puck design. The divider rails (gray) keep the panels from falling out. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/dd75b317-cf47-47c7-82d8-c279272d87b5/Electronics.PNG Sensorized Arm Cuff The force-sensing resistors (FSRs) are made from piezoresistive material that correlates the force applied on the sensing surface to the resistance of the sensor in an electrical circuit. This circuit requires the use of an operational amplifier. A printed circuit board (PCB) has been designed for the implementation of the circuit, such that each PCB accommodates four FSRs. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/246dcec0-dd48-466f-80b1-7e570cc88f4f/Calibration.PNG Sensorized Arm Cuff Calibration setup and results from a representative FSR (1s of data under each loading condition). The linear relationship has the best performance across calibration tests across 24 sensors. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/f59cdc03-00f8-4e3a-b796-1f9a3d41a686/Validation+Testbed.PNG Sensorized Arm Cuff The validation testbed with the sensorized cuff mounted on a six-axis force/torque sensor (green). Distributed forces are applied with the loading block (black). https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/ef4a519b-af50-4af7-af1e-82e4805b1d84/Validation+Cases.PNG Sensorized Arm Cuff Four different loading cases are used to validate force transmission from panel surfaces to FSRs in the sensorized cuff. Only the panels highlighted for a given case are loaded in that test. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/8e08b8b1-ea72-47fd-be0e-4379c327a16f/Validation+Results.PNG Sensorized Arm Cuff Results comparing the vertical loads measured by the force/torque sensor against the total sum of loads measured by the FSRs (in the vertical direction) for four loading cases: Single Panel, Single Column, Single Row, and Nine Panel, at three load levels. The error bars represent standard error across three repetitions. The error bars represent standard error. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/b7b57a33-a41b-433c-89cb-6db74ae3a634/Flexion+Extensions.PNG Sensorized Arm Cuff (Left) Movement performed by the subject starts at full elbow extension (transparent), going to full flexion of the elbow (opaque) and back to full extension at the end of a single repetition. (Right) Expected deformation of the arm’s muscles in response to the performed movements. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/44920d35-bd5e-4e91-a129-54c57786a3c9/Foam+Types.PNG Sensorized Arm Cuff Two types of foam are compared in this paper: polyurethane or PU (left) and blended ethylene propylene diene monomer (EPDM) (right). https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/e53add18-6542-402d-9689-dbea2f60d910/Application+Results.PNG Sensorized Arm Cuff Results from the elbow flexion–extension movements performed by the subject while wearing the Harmony exoskeleton. Active movements refer to the subject performing the elbow flexion–extension while the robot follows in the gravity-compensation mode. Passive movements refer to the robot performing the movement while the subject follows passively. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/2bd78f8d-eb47-4a9a-98bc-1d8a0458eb61/Strap+Tension+Heatmaps.PNG Sensorized Arm Cuff Heatmaps show distributed interface forces in Newtons across rows (vertical axis) and columns (horizontal axis). (Left) Distributed interface forces measured from FSR data at three strap tension levels where the third level is the tightest. (Right) Difference in distributed forces from the first level to the second level and from the second level to the third level. https://www.saadyousaf.com/hand-exoskeleton-system daily 0.75 2025-06-20 https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/afcdb487-1643-4508-8759-3eac3ff5bd8a/00_maestro_overview.jpg Maestro: Multi-DOF Hand Exoskeleton System The Maestro hand exoskeleton includes 3 finger modules (index, middle, and thumb), with 8 total DOFs actuated through Bowden cable SEAs. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/bee707f0-bbda-43d5-af83-dc722cef8a7c/01_saad_maestro_donned.jpg Maestro: Multi-DOF Hand Exoskeleton System Maestro's 8 DOFs are actuated through an EtherCAT communication architecture, enabling real-time control with position and force feedback. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/13c9d387-aa5d-4ec5-8bac-9f9fc6e9b518/01_maestro3_robot_with_electronics.png Maestro: Multi-DOF Hand Exoskeleton System Maestro's 8 DOFs are actuated through an EtherCAT communication architecture, enabling real-time control with position and force feedback. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/f25889e9-29c1-4de2-b22f-935cbeff8b70/02_maestro_full_mechatronics.jpeg Maestro: Multi-DOF Hand Exoskeleton System Bowden cable SEAs are actuated with brushed DC motors from Maxon. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/26e86e46-9f4b-4de2-95a1-b753060ce9fe/04_maestro3_electronics_box_modules.png Maestro: Multi-DOF Hand Exoskeleton System The electronics include EtherCAT motor drivers in the control module, brushed DC motors in the motion module, and power electronics in the power module. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/28a04930-9201-4656-83f4-6b353e057227/05_motor_driver.jpg Maestro: Multi-DOF Hand Exoskeleton System The EtherCAT motor drivers are the latest version supplied from Harmonic Bionics, with real-time encoder feedback to enable high-fidelity low-level position control for all motors. https://www.saadyousaf.com/instrumented-hand daily 0.75 2025-09-24 https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/b6910945-e047-436c-b4c8-ef7ed6aaab66/Instrumented+Hand+Overview.PNG Instrumented Hand The open-sourced Instrumented Hand for soft device validation measures joint level information from the thumb, index, and middle fingers. Palmar side (left) and dorsal side (right). https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/9da09c78-9465-4857-b4da-bdaae75096fd/Instrumented+Hand+Joints.PNG Instrumented Hand The Instrumented Hand with all rotary joints labeled. The CMC joint is approximated by two rotary joints, CMC1 and CMC2. The link between the thumb CMC and MCP joints is rotated to enable a more natural thumb orientation and flexion motion. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/75cbd539-bff2-4652-8f9f-3bcf7aef7433/IH+Tendon+Routing.PNG Instrumented Hand Side view of the PIP and DIP joints on the instrumented hand show the 1:1 coupling achieved by using a Kevlar braided line (1 mm, Spear-It) as highlighted in green and anchored on either end. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/62697e83-24bb-4ebd-8aff-5052b0a7083b/IH+Cross+Section.PNG Instrumented Hand Cross-sectional view of a rotary joint showing the sensor cover, magnetoresistive sensor, neodymium ring magnet, nut, bearings, shoulder bolt, and torsional spring. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/5c20749e-8610-4790-ab96-dabd1b21e6c7/IH+Joint+Side+View.PNG Instrumented Hand Side views of a rotary joint showing the indentations used to visually measure joint angles at 15 deg increments for calibration. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/2d010a30-3481-4e95-a9ca-1d8f55c75de0/IH+Joint+Calibration.PNG Instrumented Hand Calibration data from the PIP joint on the index finger with a linear regression (R^2 = 1) representative of all joints. As expected, the highly linear output for each of the joints of the Instrumented Hand supports its use in experimental validation of soft hand exoskeletons. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/def12245-ee30-4a66-afa9-445229dc675d/IH+Torque+Characterization.PNG Instrumented Hand Torque characterization results showing the relationship between torque and joint angle with the 95% confidence interval possessing a width of 0.0104 N-m at the end of the ROM. The R^2 value for the linear fit shown in red is 0.9416. Also shown is the relationship for the spring in isolation based on the manufacturer’s specifications. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/87903a45-6bce-4031-ab35-7346e7d03c3c/SPAR+Glove.PNG Instrumented Hand The SeptaPose Assistive and Rehabilitative (SPAR) Glove with individually actuated thumb, index, and secondary fingers is a soft exoskeleton which relies on the wearer’s musculoskeletal system for reaction forces. This reliance on a wearer makes the SPAR Glove a good candidate for validation with the Instrumented Hand. https://www.saadyousaf.com/maestro-kinematics-dynamics daily 0.75 2025-09-24 https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/8dbacbe0-2771-4303-99c5-5c8e0ab341bf/01_kinematics_real_links.png Maestro Kinematics and Dynamics Maestro's human-robot kinematic system defined with real linkages labeled. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/652b71dd-2b9b-47e2-9433-0a1aabec143d/02_kinematics_virtual_links.png Maestro Kinematics and Dynamics Maestro's human-robot kinematic system defined with virtual linkages labeled. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/92fbf87f-6ac3-4141-b093-1a38d9d51a93/03_cHand.PNG Maestro Kinematics and Dynamics The cHand from CHAI3D is used to visualize the estimated hand pose, based on forward kinematics from the Maestro hand exoskeleton. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/6edda505-f025-466c-9922-50a79b6f522a/04_sim_virtual_wall.png Maestro Kinematics and Dynamics Results of a virtual wall interaction in simulation, showing the implementation of the Jacobian to estimate commanded robot joint torques. The virtual wall in this example is implemented at a y-position of 20mm, at which point the commanded torques become non-zero. https://www.saadyousaf.com/dexterous-manipulation daily 0.75 2025-09-24 https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/fbdddded-b09d-44c4-8754-57d531cf332a/00_maestro_visualization_old.PNG Teleoperation for Dexterous Manipulation The Maestro hand exoskeleton with a hand visualization in the Unity engine. The forward kinematics model is used to estimate finger joint angles based on measurements from robot joint angles. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/1acc059a-b9c9-45df-a453-68f969daab2f/00_overview_unity_visualization.PNG Teleoperation for Dexterous Manipulation The virtual hand in Unity, interacting with virtual objects that impose physics interactions. The expected fingertip forces can be provided as haptic feedback through Maestro's actuation. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/1137a587-4625-4854-937d-ed3c5cec7851/01_maestro_plato_telemanipulation_block.PNG Teleoperation for Dexterous Manipulation The Maestro hand exoskeleton is used to teleoperate the remote robot hand, manipulating a block. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/dd304ff2-8948-4d79-adfb-79d2b7295152/02_maestro_plato_ball_on_block.PNG Teleoperation for Dexterous Manipulation The Maestro hand exoskeleton is used to teleoperate the remote robot end-effector fingers, whereas the paired Lambda haptic device teleoperates the end-effector position and pose. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/feac5c86-4b95-4b38-8c2e-1d3f1edf8948/03_hand_visualization_unity.PNG Teleoperation for Dexterous Manipulation A visualization of the baseline virtual hand in Unity. https://www.saadyousaf.com/wearable-hand-interfaces daily 0.75 2025-09-24 https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/d87f169e-720b-454e-9672-5866d5cda7d7/00_hand_pHRI_overview.png Wearable Hand Interfaces The proposed framework is applied to the Maestro hand exoskeleton as a case study, investigating the effects of bending stiffness and compressive stiffness at the interface on pHRI metrics. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/3d57e5b3-c581-4c6c-84bf-adff82eba869/01_simscape_model_schematic.PNG Wearable Hand Interfaces An overview of the simulation model from Simscape Multibody. The blue bar is modeled as a discretized flexible beam, corresponding to the bending stiffness properties of the dorsal plate. The highlighted spring-damper element represents the compressive stiffness at the human-robot interface, including viscoelastic properties from both human soft tissue and robot padding. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/5c58e50b-ec82-4184-a349-8de7396e3e09/02_simscape_block_diagram.png Wearable Hand Interfaces The simulation model in Simscape Multibody, developed through individual component blocks that come together to define the multi-body dynamics of the whole system. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/59bfbac9-d0fb-48d7-aa27-e13f6744e84c/03_comfort_results.PNG Wearable Hand Interfaces Results from the Bradley-Terry model, visualized as heatmaps showing the power for each interface stiffness condition across questions aggregated between the positive-negative pairs. A higher power value means that users preferred that condition more as an answer to the respective question. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/5d6abd50-7e58-41a9-b081-23506028e02c/04_interface_power_flow_schematic.PNG Wearable Hand Interfaces The power flow diagram of the system shows interface power as an unintended output. By eliminating or estimating other areas of power output in the system, we can isolate the desired interface power measurement. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/b866d6af-d3c5-4faa-9aba-d8c76c05e748/04_pHRI_performance_testbed.PNG Wearable Hand Interfaces The instrumented testbed used to measure interaction quality metrics during physical human-robot interaction with the Maestro hand exoskeleton can be operated under both a static condition (left) and a dynamic condition (right). The testbed integrates a Futek single-axis load cell and a linear encoder to measure fingertip force and displacement. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/3e08829b-5009-4fa7-aed9-6b069a9e9d01/05_performance_results.PNG Wearable Hand Interfaces Kinematic misalignment (left) and interface energy loss (right) measured for 12 participants under a static task condition, across 12 combinations of interface stiffness. The 4 levels of bending stiffness are on the horizontal axis, and the 3 levels of compressive stiffness are shown across bars. https://www.saadyousaf.com/soft-tissue-characterization daily 0.75 2025-09-24 https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/12a688e3-9a43-4109-b33c-1598cc9d1528/01_Soft+Tissue+Characterization.PNG Soft Tissue Characterization For wearable robot design, the question about where to place attachments and how user muscle activation affects interaction is tied to how these variables affect underlying soft tissue properties. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/d4847c07-3d1b-48b6-a450-0829ddd92535/02_Soft+Tissue+Characterization.PNG Soft Tissue Characterization The indenter device uses a single-axis load cell mounted on a linear carriage to measure the forcedisplacement response at the skin surface for various angles around the human forearm. The hand dynamometer measures the grip force exerted by the test subject. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/ed178f78-560e-4774-9f2f-fd7d81d908f6/03_Soft+Tissue+Characterization.PNG Soft Tissue Characterization Four sEMG sensors used in the experiment target (A) finger flexion, (B) finger extension, (C) wrist flexion, and (D) wrist extension. The indentation location along the forearm is at the midpoint between the elbow and the wrist. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/1b027c99-aba6-41cc-a912-43a471044413/04_Soft+Tissue+Characterization.PNG Soft Tissue Characterization The six angular locations measured in this experiment are shown in the forearm cross section of the right arm, as viewed from the elbow looking out towards the wrist. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/d45bd5a9-9fa7-4654-bc4c-4ab62528a6e4/06_Soft+Tissue+Characterization.PNG Soft Tissue Characterization Force-displacement response from the first indentation trial at the 180 degree angular location for one subject at four muscle activation levels along with the fitted linear regression curve. The four muscle activation levels are 0%, 15%, 30%, and 45% of maximum grip force. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/29835b44-2592-4b27-baa2-20877e09becd/07_Soft+Tissue+Characterization.PNG Soft Tissue Characterization Summary of soft tissue stiffness results from six subjects based on an average stiffness calculation from the loading section of the force-displacement response. The results from four muscle activation levels are given within each of the six angular locations. Each bar is the average of three trials and six subjects, and the error bars represent standard error. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/d8b8b84c-1c05-4ebe-a7db-c9e6240eece0/08_Soft+Tissue+Characterization.PNG Soft Tissue Characterization Summary of sEMG measurements based on an average of the four sEMG sensors during the loading section of the force-displacement response. The results from four sEMG sensors are given within each grip force level. Each bar is the average of six angular locations, three trials and six subjects, and the error bars represent standard error. https://www.saadyousaf.com/force-control-tensioning daily 0.75 2025-09-24 https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/6ca270ee-4b28-44bb-bd48-c243d9f350b5/01_cuff_wArm_isometric_human_robot.png Force Control Tensioning The proposed sensorized cuff implements force-sensing resistors (FSRs) with an actuated tensioning mechanism for measuring and modulating interface pressure. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/6bd7006e-9790-46e3-a2c5-be7dcf686c03/02_cuff_module_design_features.png Force Control Tensioning Design features highlighted for measuring interface pressure with actuated tensioning. Each FSR is housed between the (A) FSR module and the (B) force transmitter, which in turn interfaces through the (C) foam plate and (D) interface padding. The actuated tensioning mechanism is implemented through (E) cable routing over a (F) roller pin at each module. All six FSR modules are linked through (G) compliant elastomeric connectors which realize the hybrid rigid-soft cuff design. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/18c559cb-d901-409d-b078-3ee6741d5d35/03_cuff_six_panel_overview.png Force Control Tensioning The cross section of the full cuff highlights the cable routing mechanism used to evenly distribute the actuated tensioning force around the arm. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/103e908c-7f4d-451f-81de-9d225c28c0b8/04_fsr_calibration_testbed.png Force Control Tensioning The FSR calibration testbed, showing how a Futek load cell interfaces with each FSR panel through the actuated linear testbed. The controlled loading and unloading of each FSR with a ground-truth measure for force is used to calibrate the FSR voltage output. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/706848ae-cc4d-46c2-9fe8-f236d2cdeef4/05_fsr_calibration_results.png Force Control Tensioning Measured data from the FSR calibration is plotted as ground truth load cell force versus raw voltage from the FSR. A linear regression is used for the calibration. https://www.saadyousaf.com/wearable-arm-interfaces daily 0.75 2025-09-24 https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/37e50bbf-9274-4a60-8b1a-d0adfe68b166/00_arm_pHRI_overview.PNG Wearable Arm Interfaces This study explored the trade-off between user physical comfort (left) and device interaction quality (right) in arm attachment interfaces. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/2b185f2e-b568-4e75-bdc7-531e8ffd634f/01_arm_pHRI_testbed.PNG Wearable Arm Interfaces The experimental testbed for the effect of strap pretension and user muscle activation on interaction port stiffness, with (A) the actuated tensioning cuff with measurement of distributed interface forces, (B) hand dynamometer to measure user muscle activation, (C) arm anchors to ground the human, and (D) actuated linear carriage and single-axis load cell. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/e37282b3-f21c-4801-955f-6bd8e692a414/02_interaction_port_stiffness.PNG Wearable Arm Interfaces The interaction port stiffness between the human arm and the robot linkage is estimated through measurement of displacement and force at the linkage output. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/0da056ca-de19-4f82-8e63-03a655d4417e/03_results_physical_comfort.png Wearable Arm Interfaces The distributed interface force averaged across 6 FSR locations, visualized at four levels of strap pretension and across three conditions for muscle activation. The plot shows results averaged across all participants at the maximum during interaction loading. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/d1251af9-6886-46eb-854b-ed5422cdfe9c/04_results_interaction_quality.png Wearable Arm Interfaces The effective stiffness across the interaction port at the human-robot interface, visualized at four levels of strap pretension and across three conditions for muscle activation. The plot shows results averaged across all participants. https://www.saadyousaf.com/actuator-torque-sensing daily 0.75 2025-09-24 https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/b88b186b-f35c-44c0-b068-106d0049f6b2/02_joint_exploded_view.PNG Actuator Torque Sensing Exploded view of the underwater robot joint from Houston Mechatronics Inc, highlighting the harmonic drive components in the middle. For this project, we applied 4 strain gauges to the flexspline of the harmonic drive. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/3e79954b-2525-47b4-b551-1159ea5cd406/04_flexspline_FEA_w_strain_gauge.PNG Actuator Torque Sensing (Left) FEA of the flexspline, showing the equal and opposite strains imposed due to the inherent flexing in a harmonic drive. By removing the expected flexing strain from our strain gauge signals, we isolated the torque across the robot joint. (Right) The strain gauge model used in this project. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/3dfd2f23-3cc8-4c03-89cf-406fa054aa0c/02_Strain+Gauge.JPG Actuator Torque Sensing Custom application of a strain gauge on the outer surface of the flexspline, ensuring optimal contact and wiring. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/4e584809-552b-4226-9814-50906d48918b/03_flexspline_strain_gauges.PNG Actuator Torque Sensing A full set of 4 strain gauges applied to the flexspline, with adjacent pairs undergoing equal and opposite strain during no-load harmonic drive motion. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/4e66a42c-b98c-4444-b49c-5f940fc73738/04_joint_torque_testbed.PNG Actuator Torque Sensing The flexspline with strain gauges installed inside the underwater robot joint. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/a4665e32-27ca-4bb4-a4d8-2dcf556f13e9/03_poster_cropped.JPG Actuator Torque Sensing Our project presented at the Rice Engineering Design Showcase, winning the "Best Robotics Project" award. https://www.saadyousaf.com/sea-validation-control daily 0.75 2025-06-20 https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/4b7582dd-d519-4536-a044-ca2fe4334154/01_SEA_schematic.PNG Series Elastic Actuator Validation and Control The Bowden cable SEA connects the actuated motor end to Maestro's robot joint. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/dfdc030c-f9f2-42b4-a686-57aa136f2f61/02_spring_characterization_testbed.png Series Elastic Actuator Validation and Control The spring characterization testbed with a Futek single-axis load cell and a PCB actuated linear carriage, controlled through an NI USB DAQ with GUIs developed in MATLAB. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/4283a35a-7b72-4204-99ca-56b608c3b96b/04_backlash_plot.PNG Series Elastic Actuator Validation and Control Backlash calibration data from the index finger MCP joint in Maestro. As the motor is actuated through an increasing sinusoid, the robot joint follows with horizontal offsets due to Bowden cable backlash. The calibration is performed on the robot system alone without a human user. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/d192f74e-c1aa-4231-834f-7a176a2822eb/05_Interface+Power.PNG Series Elastic Actuator Validation and Control Time series data from one trial with no dorsal interface foam. The motor angle and robot joint angle are measured from Maestro's sensors. The joint torque is calculated using the SEA mechanism, with improved torque estimation through spring characterization and backlash compensation. https://www.saadyousaf.com/coursework daily 0.75 2023-01-12 https://www.saadyousaf.com/coursework/project-five-db8mh monthly 0.5 2023-01-12 https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/0ef4c93e-2f88-45db-916f-631f12412618/ASBR_01.PNG Coursework - Algorithms for Sensor-Based Robotics Yaskawa industrial manipulator used for analysis including forward kinematics, inverse kinematics, and impedance control. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/969f8e15-3c88-4cca-b73e-c2af306b58e6/ASBR_02.PNG Coursework - Algorithms for Sensor-Based Robotics Screw axes components for one arm of the Yaskawa robot which includes 7 degrees of freedom (DOFs). https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/43eae181-0af9-42da-aef4-87b0c5db20c2/ASBR_03.PNG Coursework - Algorithms for Sensor-Based Robotics Rosa Spine surgical system with eye-to-hand transformation highlighted. https://www.saadyousaf.com/coursework/project-three-crmgm monthly 0.5 2021-12-13 https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/7bd686c6-3913-4685-8c57-8c6b8ff321c6/DP_03.PNG Coursework - Dynamic Programming for Drive Cycle Results from the implementation of a linear-quadratic regulator (LQR). https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/d8e1e6a0-656e-4898-83fe-83a7b743a900/DP_04.PNG Coursework - Dynamic Programming for Drive Cycle An example drive cycle chosen for the dynamic programming problem. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/5a5c8e3c-ad57-4027-9537-bebee880c610/DP_05.PNG Coursework - Dynamic Programming for Drive Cycle Results from the implementation of dynamic programming. In most instances, it is better to not use the fuel cell power. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/ac8fb03e-475c-4050-9044-ba46ecca0bec/DP_06.PNG Coursework - Dynamic Programming for Drive Cycle Results show how the fuel cell state-of-charge (SOC) changes during the drive cycle. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/4a99e4a8-6c15-4160-9fad-d01d2b685cce/DP_02.PNG Coursework - Dynamic Programming for Drive Cycle Overview of the vehicle used for the dynamic programming problem. https://www.saadyousaf.com/coursework/project-one-6wsbx monthly 0.5 2021-12-13 https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/01e13e0b-3e27-4054-9d23-54bd096b54dc/Lathe_03.PNG Coursework - Precision Machine Design of Tabletop Lathe Overview of the tabletop lathe. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/6994da12-f55b-4cce-8895-5ad50a7d1f9f/Lathe_02.PNG Coursework - Precision Machine Design of Tabletop Lathe Aluminum cylinder machined from 1/2" to 1/4" diameter. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/f029ba3a-48ec-4999-b28c-e09685c7ae78/Lathe_01.PNG Coursework - Precision Machine Design of Tabletop Lathe Dial to control radial depth of the tool tip mounted on the carriage which travels in the axial direction. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/fb1474ab-cccd-48e1-8c90-4cd2b4939924/Lathe_04.PNG Coursework - Precision Machine Design of Tabletop Lathe The radial depth of the tool tip is controlled with a flexure. A DC motor powers the lathe. https://images.squarespace-cdn.com/content/v1/606ba05a20527b40bfd626d3/5a9b1daa-30bd-4ee9-a4fa-389d4cf182b7/Lathe_06.PNG Coursework - Precision Machine Design of Tabletop Lathe Motor side mount.