Hardware & control
Taxel
A taxel is one spatial sensing element in a tactile array, analogous to a pixel in an image sensor. A taxel may measure pressure, force, deformation, capacitance, resistance, or another contact-related quantity at a defined location; its value is meaningful only with the sensor geometry, calibration, units, range, and timing.
Also known as: tactile pixel, tactile sensing element, tactile sensor element
Updated
One taxel is one sample location, not a complete touch image
A tactile array distributes sensing elements across a fingertip, palm, gripper, skin panel, or glove. Each taxel produces a local reading. The array becomes a spatial tactile observation only when software retains where those readings sit on the physical surface.
Taxel count is therefore not the same as spatial resolution. Element size, pitch, inactive gaps, surface curvature, wiring, cross-talk, interpolation, and the contact material determine which patterns the sensor can distinguish. Two arrays with the same count can cover different areas and measure different physical quantities.
Geometry and calibration give a taxel meaning
Tactile-sensing data should identify each taxel's location or region, units, zero point, usable range, sample rate, calibration method, and validity state. Saturation is different from a large valid force. A disconnected or masked element is different from zero contact.
When a sensor is remounted, its taxel-to-body transform may change. A learning system that uses 3D taxel positions needs the coordinate frame and time-varying hand pose as well as the pressure values. A flattened vector without layout metadata discards the spatial structure that made the array useful.
Flexion and wear can create misleading readings
Wearable tactile gloves illustrate the problem. Spark-0 reports 460 sensing points per HumanTouch glove but masks 140 over flexion regions because bending can create spurious pressure. The remaining 320 taxels are grouped into palm and finger regions and used for the model observation and contact-map construction.
The example shows why “number of sensors” is not enough for a dataset card. Useful tactile records preserve the raw layout, health mask, filtering rules, calibration, remounting history, and reasons that any taxels were excluded.
Sources
Related terms
Hardware & control
Tactile sensing
Tactile sensing is the detection and measurement of physical contact properties at a robot's surface or contact interface. Depending on the sensor, it can report pressure or force distribution, contact location, shear, vibration, slip, texture, temperature, or deformation. Tactile data complements vision by measuring interactions that may be hidden at the point of contact.
Hardware & control
Force–torque sensing
Force–torque sensing is the measurement of forces and rotational moments transmitted through a robot interface, commonly with a six-axis sensor reporting three force components and three torque components. The combined quantity is a wrench expressed at a specified origin and in a specified coordinate frame. It measures the resultant load at the sensor, not a spatial pressure map across every contact.
Data & collection
Sensor calibration
Sensor calibration is the estimation and documentation of parameters that map raw sensor readings into physically meaningful values and known spatial relationships. In robotics it can include scale, bias, distortion, intrinsic camera parameters, sensor-to-sensor or sensor-to-robot transforms, and timing offsets. Calibration does not remove all noise or drift.
Data & collection
Data synchronisation
Data synchronisation is the process of placing sensor, state, action, annotation, and outcome records on a common timeline so samples that describe the same physical instant or transition can be matched. It requires trustworthy timestamps or trigger relationships and an explicit rule for handling streams with different rates, delays, dropped samples, and clock offsets.
Data & collection
Robot state
Robot state is the set of variables used to describe a robot at a particular time, such as joint positions and velocities, base pose, end-effector pose, gripper state, actuator measurements, or estimated motion. In control theory, a complete state contains enough information to predict future evolution given an action; in robot datasets, “state” often means only the measured or estimated subset that was logged.