A
Actuator
The motor + gear assembly that moves a robot's joints. Think of it as the "muscle." Electric actuators (most common) vs hydraulic (rare in homes due to complexity).
Autonomy
The robot's ability to perceive, plan, and execute tasks without human intervention. See: Teleoperation vs Autonomy.
B
Bipedal
Two-legged locomotion. Humanoid robots are bipedal (like humans) vs quadrupedal (four legs, like Spot by Boston Dynamics).
BotQ
Figure's manufacturing philosophy focusing on high-volume, low-cost production methods (injection molding, die casting) to scale humanoid robots affordably.
C
Collision Detection
Sensors (cameras, LiDAR, tactile) that identify obstacles and stop or redirect the robot before impact. Critical for homes with pets and children.
Company-Stated
Dates, specs, or capabilities announced by the company but not independently verified. See: Our living timeline for confidence labels.
D
Degrees of Freedom (DoF)
Number of independent movements a joint can make. A human arm has ~7 DoF (shoulder 3, elbow 1, wrist 3). More DoF = more dexterous, but also more complex and expensive.
Depth Sensor
Camera that measures distance to objects (e.g., RealSense, LiDAR, structured light). Used for 3D mapping and collision avoidance.
E
E-Stop (Emergency Stop)
Physical button and/or app control that immediately halts all robot motion. Essential safety feature. See: Safety 101.
Expert Mode
1X's branding for teleoperation—human operators remotely control the robot when autonomy isn't sufficient.
G
Geofencing
Virtual boundaries that prevent the robot from entering specific zones (e.g., stairs, nursery, outside). Set via app or room mapping.
Grasping
The ability to securely hold and manipulate objects. Requires coordinated finger control, force sensing, and object recognition. Failure modes: slippage, crushing fragile items.
I
Imitation Learning
Training method where robots learn by watching humans (or teleoperators) perform tasks, then replicate those behaviors. Key to scaling autonomy.
IMU (Inertial Measurement Unit)
Sensor that measures acceleration, rotation, and orientation. Used for balance and fall detection in bipedal robots.
L
LiDAR
Light Detection and Ranging. Laser-based sensor that maps 3D environments by measuring distance. More accurate than cameras in low light; used in navigation and obstacle avoidance.
Locomotion
The robot's ability to move (walk, balance, climb). Bipedal locomotion is harder than wheeled mobility but necessary for stairs and human-centric spaces.
M
Manipulation
Using arms/hands to interact with objects—picking, placing, grasping, twisting. Requires coordinating vision, force control, and motion planning.
Mapping (SLAM)
Simultaneous Localization and Mapping. The robot builds a 3D map of your home while tracking its own position within that map.
O
On-Device vs Cloud Compute
On-device: AI models run on the robot's onboard computer (faster, more private). Cloud: models run on remote servers (more powerful, but requires internet and raises privacy concerns).
P
Payload
Maximum weight the robot can lift safely. Humanoids typically handle 5–25 kg (11–55 lbs)—enough for dishes, laundry, light groceries.
Perception
The robot's ability to "see" and interpret its environment—object recognition, depth estimation, scene understanding. Combines cameras, depth sensors, and AI models.
Policy (AI)
A trained model that maps observations (what the robot sees) to actions (what it should do next). Example: "If I see a plate on the counter, grasp it and move it to the dishwasher."
S
SLAM (Simultaneous Localization and Mapping)
Algorithm that lets the robot map its environment while tracking its own position. Essential for navigation in homes.
Subscription Model
Monthly payment for robot access without owning it. Typically includes support, updates, and teleop minutes. See: Price Models Explained.
T
Tactile Sensing
Force/pressure sensors in fingers/hands that detect contact and grip strength. Prevents crushing fragile objects or dropping slippery items.
Teleoperation (Teleop)
Human remotely controls the robot in real time. Used as fallback when autonomy fails. See: Full explainer.
Torque
Rotational force applied by actuators. Higher torque = stronger joints, but also higher injury risk if not force-limited.
V
Vision-Language Model (VLM)
AI model that understands both images and text. Allows robots to respond to natural language commands like "load the dishwasher" while interpreting visual scenes.
Z
ZMP (Zero Moment Point)
Control theory concept for bipedal balance. The point on the ground where the sum of all forces equals zero—if ZMP stays within the robot's foot polygon, it won't tip over.