Series Elastic Actuator (SEA)

The defining mechanical feature is a compliant element (typically a linear or torsional spring) placed in series between the motor's output and the actuator's end-effector or load. This creates a deliberate, measurable deflection under force.

• Key Benefit: This physical spring acts as a built-in force sensor. By measuring the spring's deflection (via an encoder or strain gauge), the actuator can directly and accurately infer the output force using Hooke's Law (Force = Spring Constant × Deflection).

SEAs excel at accurate and responsive force/torque control. The direct spring deflection measurement provides a low-noise, low-latency force signal, bypassing the need to estimate force from noisy motor current readings in stiff actuators.

• Applications: This enables gentle physical interaction, precise impedance modulation, and stable contact with uncertain environments. It is critical for tasks like assembly, collaborative robotics (cobots), and legged locomotion where ground reaction forces must be carefully managed.

The series spring provides intrinsic mechanical compliance, which offers two major advantages:

• Impact Protection: It passively absorbs and filters high-frequency shock loads (e.g., foot strikes in walking robots, unexpected collisions). This protects both the delicate gearbox/motor and the external environment.
• Inherent Safety: In human-robot interaction, the spring yields under excessive force, reducing peak impact forces and the risk of injury, making SEAs a preferred choice for physical human-robot interaction (pHRI).

The elastic element can temporarily store and release mechanical energy, enabling dynamic, efficient motions.

• Cyclic Tasks: In activities like running or jumping, energy can be stored in the spring during landing and released to assist propulsion, mimicking biological tendons. This can reduce peak motor power requirements and improve overall system efficiency.
• Resonance Exploitation: Controllers can be designed to exploit the system's natural resonant frequency for periodic tasks, further reducing energy consumption.

SEAs are the hardware embodiment of impedance control, where the actuator is programmed to exhibit a desired dynamic relationship between motion and force (like a programmable mass-spring-damper).

• Impedance Control: The controller commands force based on measured position/velocity error. The SEA's physical spring makes implementing this stable and straightforward.
• Contrast with Stiff Actuators: Traditional rigid actuators often implement admittance control, where they measure force and command motion. SEAs implement impedance control more naturally and robustly.

The primary design trade-off involves the spring constant (stiffness).

• Softer Spring: Provides better force fidelity, shock absorption, and safety but limits bandwidth (maximum frequency of force control) and can introduce oscillation.
• Stiffer Spring: Increases force control bandwidth and positional accuracy but reduces the benefits of compliance and force resolution.

Designers must select spring stiffness to match the dominant task requirements—high-bandwidth precision positioning versus safe, gentle interaction.

A robot control strategy that regulates the dynamic relationship between force and motion at the end-effector. Instead of directly commanding position or force, it makes the robot behave as a programmable mass-spring-damper system. This is a control paradigm, whereas a Series Elastic Actuator is a hardware implementation that naturally exhibits compliant behavior, making it an ideal physical platform for implementing impedance control. Key characteristics:

• Stiffness, Damping, Inertia: The three parameters that define the desired dynamic behavior.
• Interaction Stability: Often provides better stability during unexpected contact than pure position control.
• Versatility: The same controller can execute delicate insertions and rigid positional tasks by adjusting parameters.

A robot control strategy where measured external forces (e.g., from a force-torque sensor) are used to compute a desired motion deviation. It is the dual of Impedance Control. In an admittance control architecture, a high-gain position controller is typically used, and the force measurements modify the position command. Comparison with SEA:

• SEA as Admittance: An SEA can simplify admittance control because its built-in spring provides a direct, high-fidelity force measurement via deflection, eliminating the need for a separate wrist-mounted force-torque sensor.
• Control Loop: Admittance control uses a force-in, position-out structure.
• Use Case: Ideal for applications like physical human-robot interaction or cooperative carrying, where the robot should 'give way' to applied forces.

A condition in robotic grasping where the set of contact forces applied by the gripper can generate any resultant wrench (combined force and torque) on the object. This ensures the object can be held securely against arbitrary external disturbances. Relationship to SEA:

• Force Modulation: An SEA's precise force control allows a gripper to actively regulate contact forces to maintain force closure, even if the object shifts or external forces are applied.
• Slip Prevention: By sensing and controlling grip force via the actuator's spring deflection, an SEA can prevent slip, which is a failure of force closure.
• Analysis: Force closure is a theoretical property analyzed using the grasp wrench space to evaluate grasp quality.

A robot's ability to sense its own internal state—such as joint position, velocity, motor current, and link torque—without external references. The Series Elastic Actuator fundamentally enhances proprioception.

• High-Fidelity Force Sensing: The spring deflection in an SEA provides a direct, joint-level measurement of output torque, which is a key proprioceptive signal often missing in rigid actuators.
• Collision Detection: The sensitive force sensing enables robots to detect unexpected contact (e.g., collisions with humans or the environment) purely from internal sensor data.
• State Estimation: Combines motor encoder data (position before the spring) and spring deflection to estimate both joint state and interaction forces.

An advanced control method where a dynamic model of the system is used to predict future behavior over a finite horizon. It then optimizes a sequence of control inputs to minimize a cost function, applying only the first step before re-solving. Relevance to SEA:

• Managing Compliance: MPC can explicitly plan around the dynamics introduced by the SEA's spring, optimizing for smooth force trajectories and stable contact transitions.
• Constraint Handling: Can incorporate constraints on maximum spring deflection (force), motor torque, and velocity, which are critical for SEA safety and performance.
• High Performance: Used in advanced robotic locomotion and manipulation to achieve dynamic, force-rich tasks that benefit from an SEA's hardware capabilities.

A fundamental control technique that calculates and commands the joint torques needed to counteract the weight of a robot's own links. This allows the arm to move as if in zero gravity, making it easier to control and safer for interaction.

• SEA Implementation: In an SEA-based robot, gravity compensation torques are commanded based on the robot's kinematic model. The SEA's force control loop then accurately delivers these torques at the joint.
• Backdrivability: Effective gravity compensation, combined with an SEA's low reflected inertia, makes a robot highly backdrivable. A human can easily move the robot's limbs, which is crucial for collaborative applications and teaching by demonstration.
• Foundation for Interaction: It is often the first layer in a hierarchical controller, upon which task-specific force or impedance behaviors are layered.