Tendon vs. Actuator: Toward a Hybrid Approach

As we move beyond v0.1, questions of how limbs should move become central. Robotics teams worldwide are testing two main strategies:

• Direct Actuation – electric motors or hydraulic actuators placed at or near the joint.

• Tendon-Driven Systems – motors or pulleys pull cables routed along the limb, mimicking muscles and tendons.

Direct Actuation: Power and Precision

• Strengths: High torque, direct power transfer, simpler design. Ideal for hips, shoulders, and major load-bearing joints.

• Weaknesses: Hard mechanical impacts, less compliant with unexpected forces, produces the “careful walking” effect as controllers manage momentum.

Tendon-Driven: Mimicking Biology

• Strengths: Soft, compliant motion. Energy can be stored elastically, making movement more fluid and natural. Safer for human interaction, lighter at the extremities.

• Weaknesses: Complex routing of cables, bouncy dynamics, higher maintenance from wear and slack. Space requirements can make compact designs challenging.

Hybrid Approach: Strength + Finesse

Our intuition—and echoed by several robotics leaders—is that a hybrid path may offer the best near-term solution:

• Actuators at primary power points (hips, shoulders, torso) for stability and speed.

• Tendons in intermediate and endpoint regions (biceps → forearms, calves → feet, and hands) for humanlike finesse and compliant interaction.

This balance mirrors the human body: large muscle groups provide raw force, while tendons and smaller muscles refine movement and absorb shock.

Why This Matters for ELA

For Project ELA, early prototypes (v0.1 → v0.3) will not immediately require tendon-driven design. But the principle informs long-term architecture:

• Safety – ELA must be both powerful and safe in shared environments.

• Energy Efficiency – tendon systems recycle some energy during motion.

• Embodiment Goals – natural movement fosters presence and believability.

Future Watchpoints

• 1X’s Neo/Eve – tendon-driven, compliant, highly humanlike, but complex  .

• Boston Dynamics / Apptronik – largely actuator-bas

  ed, robust, powerful, slower to embody “softness.”

• Academic Labs – tendon research often leads to breakthroughs in wearable exoskeletons and prosthetics, which could spill over into robotics.

See It in Motion

1X’s Eve / Neo (Tendon-Driven, Compliant Movement)

1X – Humanoid Walking Demo

Boston Dynamics’ Atlas (Hydraulic Actuators, Direct Power)

Atlas – Parkour and Dynamic Motion

Apptronik’s Apollo (Electric Actuators, Smooth but Careful)

Apptronik – Apollo Introduction

Media & Methods Lab UTL ETH Zurich

The Design and Fabrication of a Soft Robotic Hand

References

1. Pratt, J., & Collins, S. (2000). Mechanical Design and Control of a Walking Robot Using Direct Actuation. MIT Leg Lab Report.

2. Tonietti, G., Schiavi, R., & Bicchi, A. (2005). Design and Control of a Variable Stiffness Actuator for Safe and Fast Physical Human/Robot Interaction. IEEE International Conference on Robotics and Automation.

3. Vanderborght, B. et al. (2013). Variable Impedance Actuators: A Review. Robotics and Autonomous Systems, 61(12), 1601–1614.

4. Alexander, R. McNeill. (1991). Elastic Mechanisms in Animal Movement. Cambridge University Press.

5. 1X Technologies. (2023). Introducing Eve and Neo: Human-Safe Humanoids. Company Whitepaper.

6. Dollar, A. M., & Herr, H. (2008). Lower Extremity Exoskeletons and Active Orthoses: Challenges and State of the Art. IEEE Transactions on Robotics, 24(1), 144–158.