In robotics, where a motor is placed is often as important as what the motor does. Traditional robotic arms and legs are "serial" chains: each joint carries the weight of every motor further down the limb. This creates a "distal mass penalty"—a compounding cycle where distal weight requires heavier proximal motors, which in turn increases the total inertia of the system.
A growing body of research suggests that the key to achieving human-like agility and safety is proximal actuation: relocating heavy motors away from the extremities and toward the robot's "core" (the torso or base). By decoupling the mass of the actuator from the motion of the joint, researchers are building robots that are faster, more efficient, and inherently safer.
The Distal Mass Penalty
To understand why proximal actuation works, we must first look at why traditional designs fail. When a motor is placed directly at a robot's "elbow" or "knee," it adds to the rotational inertia of the entire limb. As Beomjoon Kim and colleagues note in Design of a low-cost and lightweight 6 DoF bimanual arm for dynamic and contact-rich manipulation, high distal mass creates a hardware bottleneck. It limits the maximum end-effector speed and increases the energy required for every movement.
In the context of wearable robotics, this penalty becomes a biological burden. Shuo Wang and Ziran Zhang explain that placing heavy components at the extremities of an exoskeleton increases the "metabolic cost" of walking. The user must expend more energy just to swing the added weight of the device, often negating the assistance the robot was intended to provide.
Decoupling Mass from Motion
The solution is to move the "muscles" (actuators) to the "body" (base or torso) and transmit power through lightweight mechanisms. Several strategies have emerged to achieve this:
- Tendon and Cable Drives: Systems like the iCub use cable-driven transmissions to relocate motors to the torso, allowing for slim, anthropomorphic limbs with minimal inertia. Similarly, Hyunchul Shim and David Hyunchul Shim utilized Bowden cables to create a 236g hand capable of lifting 25kg, effectively moving the "weight" of the grip to the robot's torso.
- Linkage Systems: Instead of cables, some robots use rigid linkages. The AGILOped humanoid, developed by Grzegorz Ficht, uses parallel kinematic chains to concentrate mass near the torso. This allows the robot to perform explosive movements like jumping because the legs themselves are incredibly light.
- Chain-and-Sprocket: In quadrupedal research, Aman Singh and Shishir N. Y. Kolathaya demonstrated that moving knee actuators to the hip joint via a chain drive enables faster swing phases and more responsive locomotion.
Speed through Low Reflected Inertia
The most striking benefit of proximal actuation is the increase in dynamic performance. When the moving mass of an arm is reduced, it can accelerate at rates that would destroy a traditional robot.
Simon Guist and the team behind PAMY2 Safe & Accurate at Speed with Tendons achieved end-effector speeds nearly triple those of conventional collaborative robots. By moving heavy pneumatic actuators to a fixed base, they minimized "reflected inertia"—the resistance felt at the joint due to the motor's own mass and gearing.
This reduction in inertia also solves a critical safety paradox: usually, higher speed means higher danger. However, because a proximally actuated arm has very little moving mass, its kinetic energy during a collision is significantly lower. This allows robots to perform high-speed tasks, like table tennis or "ballistic" hitting, while remaining inherently safe for human interaction.
Predictable Dynamics and Efficiency
Beyond speed, proximal mass distribution simplifies the "math" of walking and reaching. Yan Ning and colleagues found that concentrating mass near the center of mass (CoM) in the NING Humanoid made leg dynamics more predictable. When the limbs are light, their movement doesn't "yank" the body out of balance, facilitating more stable high-speed locomotion.
This mirrors biological efficiency. Tom Van Wouwe and Scott L. Delp showed through musculoskeletal simulation that humans with higher lower-leg-to-thigh length ratios (lighter extremities) are more efficient runners. Even elite pianists, as studied by Shinichi Furuya, achieve extreme speeds by using proximal joints (the elbow and shoulder) to drive distal movements, rather than relying solely on small finger muscles.
By shifting the mechanical workload to the core, proximal actuation allows robots to transcend the limitations of their own weight, moving with a level of grace and explosiveness previously reserved for biological systems.