Tendon-Driven Robotic Finger Grasp Performance

ISEF Category: Robotics and Intelligent Machines

Subcategory: Biomechanics  ·  Difficulty: Advanced  ·  Setup: University Lab  ·  Time: Full Year

The Hook

A human hand can grab a mug, a pen, and a banana with the same fingers. A stiff robot finger usually cannot. That gap is what this project tackles. You will test whether a tendon-driven design can beat a rigid-link baseline on awkward, off-center objects.

What Is It?

This project studies a robotic finger that uses tendons, which are cables that pull like human tendons, instead of only stiff joints. In the human hand, flexor tendons share load and let fingers wrap around odd shapes. A differential pulley system spreads motion across the finger, so one motor can help the finger adapt when the object is not centered.

Think of it like a backpack with one strap versus two. A rigid finger is like one stiff strap. It works in one position, then fails when the load shifts. A tendon-driven finger can move a little on its own as the object pushes back. That extra motion can improve contact, stability, and grasp success on round, flat, or lopsided objects.

Why This Is a Good Topic

This is a strong science fair topic because you can test one clear idea, adaptive tendon routing helps grasp irregular objects better than a rigid-link design. You can measure success, contact location, and object eccentricity with repeatable trials, so the data are easy to compare. The project connects to prosthetics, assistive devices, and soft robotics, which gives it real-world value. You can also learn design iteration, control tuning, and statistical analysis.

Research Questions

• How does object eccentricity affect grasp success in a tendon-driven finger compared with a rigid-link baseline?
• What is the effect of object shape, such as round, oval, or off-center, on the force needed to hold a grasp?
• Does a single-motor tendon system maintain grip across a wider range of object sizes than a rigid finger?
• To what extent does tendon routing geometry change the finger’s ability to self-adapt around irregular objects?
• Which contact points on the object most strongly predict successful grasp retention?
• How does the tendon-driven design compare with the rigid-link baseline in repeatability across many trials?

Basic Materials

• 3D-printed finger parts or access to a printer
• Small DC or servo motor with controller
• Fishing line, nylon cord, or thin tendon cable
• Assorted household objects with different shapes and sizes
• Digital kitchen scale or force gauge
• Phone camera for recording trials
• Ruler or calipers for measuring object dimensions
• Clamp stand or simple test frame
• Tape and zip ties for mounting
• Spreadsheet software for logging results.

Advanced Materials

• 3D-printed finger prototypes with interchangeable pulley layouts
• Torque-controlled motor or servo test bench
• Load cell with data acquisition system
• Motion capture markers or high-speed video setup
• Force plate or pressure film for contact mapping
• Set of standardized test objects with known eccentricity
• CAD software for mechanism iteration
• Bench power supply
• Encoder for joint angle tracking
• Materials for comparison fingers with different stiffness profiles.

Software & Tools

• Python: Organizes trial data, computes success rates, and runs statistical tests.
• ImageJ: Measures object dimensions, contact region size, and frame-by-frame motion from video.
• Fusion 360: Helps you model finger geometry, pulley paths, and comparison prototypes.
• GeoGebra: Lets you plot grasp success against eccentricity and inspect trends quickly.
• RStudio: Helps you run regression models, confidence intervals, and significance tests.

Experiment Steps

1. Define one performance metric first, such as successful hold, slip time, or peak grip force.
2. Choose the object feature you will vary, such as eccentricity, size, or surface curvature.
3. Build a baseline finger and a tendon-driven version with the same outer dimensions.
4. Plan a fair test setup that keeps object placement, motor input, and mounting position consistent.
5. Design a data table that records both outcome data and motion data for every trial.
6. Decide which statistical test will compare the two designs and answer your research question.

Common Pitfalls

• Changing the object’s starting position between trials, which makes eccentricity data impossible to compare.
• Comparing two finger designs with different outer sizes, which confounds geometry with mechanism type.
• Measuring success by eye only, which hides partial slips and weak grasps.
• Ignoring tendon stretch or backlash, which makes the finger behave differently after repeated trials.
• Testing too many object shapes at once, which spreads your sample size too thin to support clear conclusions.

What Makes This Competitive

A competitive version goes beyond asking whether the tendon finger works. You compare it against a fair baseline, define a sharp metric, and use enough trials to show a real effect. Strong projects also test why it works, not just whether it works, by linking performance to contact location, eccentricity, or tendon routing. A deeper analysis, like regression or failure mode classification, can turn a good build into a serious engineering study.

Project Variations

• Test the same finger on foam, plastic, and glass objects to see how surface friction changes grasp success.
• Swap the single tendon routing for two different pulley geometries and compare adaptive wrapping performance.
• Add soft fingertip pads and measure whether compliance improves eccentric grasp retention more than tendon routing alone.

Learn More

• NASA NTRS: Search for papers on robotic grippers, underactuated hands, and tendon-driven mechanisms in the NASA technical reports database.
• PubMed: Search review articles on human hand biomechanics and tendon mechanics to connect biology with design.
• The Open Source Leg documentation: Read the free engineering notes on compliant mechanisms and force transmission for design ideas.
• MIT OpenCourseWare: Look for robotics and mechanics courses with lecture notes on actuation, kinematics, and grasping.
• IEEE Xplore: Search for peer-reviewed papers on underactuated hands, tendon routing, and adaptive grasping, and use abstracts and author manuscripts when available.