M
Motion feedback
Position and velocity measurements allow the controller to compare commanded movement with the robot’s actual movement.
Robotics Sensing and Feedback Engineering Guide
Robot Sensors: Types, Performance and Selection
An engineering guide to position, force, tactile, range, inertial and safety sensing in robotic systems
Robot sensors measure the robot’s internal state, its interaction with objects and the conditions in its environment.
This guide explains the principal sensor categories used in robotic systems, how their measurements reach the controller, and which performance, calibration and integration parameters engineers should evaluate.
What Is a Robot Sensor?
A robot sensor is a device or sensing subsystem that measures a physical quantity and converts it into information the robotic system can use.
Robot sensors may measure joint position, speed, force, torque, contact, distance, acceleration, temperature, current or environmental conditions.
The measurement is normally processed, calibrated and transmitted to a controller, estimator or safety system. The useful output therefore depends on the complete sensing chain, not only the sensing element.
In summary
- The sensing element responds to a physical quantity.
- Signal conditioning converts the response into usable data.
- Calibration relates the data to known units or reference conditions.
- Communication transfers the reading to the robot system.
- Timestamps identify when the measurement was valid.
- Software interprets the measurement in the correct coordinate frame.
- Control or estimation uses the result to update robot behaviour.
Why Do Robots Need Sensors?
F
Interaction control
Force, torque and tactile measurements allow the robot to detect contact and regulate its interaction with workpieces, tools or people.
E
Environment detection
Range, proximity, LiDAR and vision systems allow the robot to detect objects, surfaces, obstacles and workspace conditions.
S
State estimation
Measurements from encoders, IMUs, range sensors and other devices can be combined to estimate position, orientation, velocity and other robot states.
C
Condition monitoring
Current, temperature, vibration and diagnostic sensors can expose overload, wear, abnormal operation or changing environmental conditions.
P
Safety-related detection
Safety-rated sensing systems may detect people or protected-zone intrusion as part of a defined safety-related control system.
Sensing provides evidence about the robot and its environment. The controller still requires a model, decision logic and defined response before that evidence changes robot behaviour.
Internal-State vs External-State Sensors
| Internal-state sensing | External-state sensing |
|---|---|
| Measures the robot itself | Measures the environment or interaction |
| Joint position | Object distance |
| Motor speed | Contact force |
| Motor current | Surface pressure |
| Joint torque | Obstacle position |
| Temperature | Light or environmental conditions |
| Battery state | Human presence |
| IMU orientation and acceleration | Workpiece or terrain state |
Internal sensing supports robot proprioception: an estimate of the robot’s own configuration and motion. External sensing supports perception and interaction: an estimate of what is around the robot or what forces act between the robot and its environment.
Some sensors cross the boundary. A wrist force-torque sensor, for example, measures an interaction force through a sensor installed within the robot.
explicitly defines external-state sensing around environmental state and interaction with the environment.
How Does a Robot Sensor Work?
- A physical quantity changes.
- A sensing element responds to the change.
- Signal-conditioning electronics amplify, filter or digitize the response.
- Calibration converts the raw signal into a defined physical measurement.
- The system attaches units, a timestamp and a coordinate frame where needed.
- Communication transfers the reading to the controller or computer.
- Filtering or estimation combines the reading with a robot model or other measurements.
- The robot uses the resulting information for control, monitoring, planning or safety.
Sensor signal chain
- Physical quantity
- Sensing element
- Signal conditioning
- Calibration
- Timestamp and coordinate frame
- Communication
- Filtering or state estimation
- Robot decision or control
What Types of Sensors Are Used in Robotics?
These categories overlap. A single sensor may support more than one system function. An encoder may support motor commutation, position control and safety-related monitoring. An IMU may support stabilization, state estimation and fault detection.
1. Position and velocity
Encoders, resolvers, Hall sensors and limit switches for joint and motor feedback.
2. Force and torque
Six-axis wrist sensors, joint torque sensing and load cells for interaction control.
3. Tactile and touch
Pressure arrays, capacitive skins and elastomer sensors for local contact.
4. Proximity and distance
Inductive, capacitive, ultrasonic, infrared and time-of-flight sensors.
5. LiDAR and laser range
2D and 3D laser scanners for mapping, localization and obstacle detection.
6. Inertial sensors and IMUs
Accelerometers, gyroscopes and optional magnetometers for motion estimation.
7. Condition and environmental
Current, temperature, vibration, humidity and battery-state monitoring.
8. Safety-related sensing
Safety laser scanners, light curtains and safety-rated encoders.
Machine vision is an adjacent category covered in the machine-vision cameras, optics, lighting and calibration guide.
How Do Robots Measure Joint Position and Speed?
Position sensors measure the location or displacement of a robot joint, motor shaft or linear axis. Velocity may be measured directly, or calculated from position change over time. The appropriate sensor depends on the required accuracy, speed, environment, startup behaviour and integration architecture.
Common technologies include rotary encoders, linear encoders, resolvers, Hall sensors, potentiometers and limit switches. Feedback may be installed on the motor shaft or at the joint output.
Absolute vs incremental encoders
An absolute encoder reports a defined position value after startup without requiring the system to count movement from an unknown initial position. An incremental encoder reports changes in position through pulses or signal cycles. The controller normally establishes an absolute reference through a reference mark, homing procedure or separate position source.
| Motor-side encoder | Output-side encoder |
|---|---|
| Measures the motor or reducer input | Measures the actual joint output |
| Supports motor commutation and servo control | Exposes gearbox and structural errors |
| Cannot directly observe gearbox lost motion | Measures the position after the transmission |
| Usually easier to integrate | Requires joint-side packaging and calibration |
Output-side measurement can expose gearbox elasticity, reversal error and transmission effects that a motor-side encoder cannot observe directly.
See servo-motor encoder selection and harmonic reducer integration for joint-level feedback context.
How Do Robots Measure Force and Torque?
Force and torque sensors measure mechanical loading between parts of the robot or between the robot and its environment. A six-axis force-torque sensor reports three orthogonal force components and three torque components: Fx, Fy, Fz, Tx, Ty and Tz.
Common sensor locations: Robot wrist, end effector, joint output, robot base, tool fixture, assembly station and gripper finger.
Common applications: Insertion and assembly, grinding and polishing, contact detection, force-controlled manipulation, process monitoring, payload estimation, human-robot interaction and quality measurement.
Critical parameters
- Force range, torque range, resolution, accuracy and repeatability
- Cross-axis coupling, overload capacity, stiffness and bandwidth
- Temperature drift, noise and interface
A high measurement range can reduce sensitivity to small forces. A high overload rating does not establish measurement accuracy after an overload event. Force-control performance also depends on sensor location, structural stiffness, signal filtering, latency and controller bandwidth.
What Are Tactile Sensors in Robotics?
A tactile sensor measures contact conditions over a point or area of the robot’s surface. Depending on its architecture, it may measure normal pressure, shear force, contact location, contact shape, slip, vibration, texture or temperature.
Tactile architectures: Resistive, capacitive, piezoelectric, piezoresistive, optical, magnetic, fluidic and elastomer-based.
Typical applications: Gripper fingers, robot hands, soft grippers, robot skin, contact localization, slip detection, object classification and human interaction.
Critical parameters: Sensing area, spatial resolution, pressure or force range, normal and shear sensitivity, response time, hysteresis, drift, durability, surface compliance and replaceability.
A wrist force-torque sensor measures the combined load transmitted through the robot wrist. A tactile array measures local contact distribution at the robot’s surface. The two can complement each other but are not interchangeable.
How Do Robots Detect Nearby Objects and Distance?
| Sensor type | Operating principle | Typical strength | Main limitation |
|---|---|---|---|
| Inductive proximity | Detects conductive targets electromagnetically | Robust metal detection | Limited to conductive material |
| Capacitive proximity | Detects changes in electric field | Can detect non-metal materials | Sensitive to environment and setup |
| Ultrasonic | Measures sound time of flight | Low cost and useful in poor lighting | Beam width, reflections and material effects |
| Infrared proximity | Measures emitted or reflected infrared energy | Compact and inexpensive | Surface and ambient-light dependence |
| Time-of-flight optical | Measures light travel time or phase | Compact distance measurement | Reflectivity, sunlight and multipath effects |
| Laser triangulation | Measures geometry of reflected laser spot | High precision over defined range | Alignment and surface dependence |
| Radar | Measures reflected radio-frequency energy | Robust in some dust, fog or lighting conditions | Resolution and integration trade-offs |
| Mechanical switch | Detects physical contact | Simple and deterministic | Requires contact |
No distance sensor works equally well for every object and environment. The correct choice depends on target material, colour, geometry, distance, field of view, lighting, dust, water, cross-talk and required response time.
What Is LiDAR Used for in Robotics?
LiDAR measures distance by transmitting laser light and analysing the returned signal. Robotic systems may use two-dimensional LiDAR for planar obstacle detection or three-dimensional LiDAR for mapping, localization, navigation and environment reconstruction.
Critical parameters
- Minimum and maximum range, range accuracy, angular resolution
- Field of view, scan frequency, point rate and return intensity
- Multi-return capability, latency, synchronization
- Sunlight performance, environmental rating and laser classification
Point count is not the same as useful environmental resolution. The usable result also depends on angular distribution, target reflectivity, motion distortion, calibration, filtering and the robot’s operating range.
Where laser products are integrated, laser-radiation classification and product safety may fall under .
What Does an IMU Measure in a Robot?
An inertial measurement unit normally measures linear acceleration and angular velocity. Some IMUs also include a magnetometer or an internal orientation estimate. The IMU does not directly measure global position.
Typical elements: Three-axis accelerometer, three-axis gyroscope, optional magnetometer, temperature sensing, signal processing and internal calibration.
Key parameters
- Accelerometer and gyroscope range, noise density, bias and bias instability
- Scale-factor error, cross-axis sensitivity, sampling rate and bandwidth
- Latency, temperature drift and vibration sensitivity
Integrating acceleration to estimate velocity and position accumulates error. IMUs are therefore commonly combined with encoders, vision, LiDAR, GNSS or other reference measurements.
Which Sensors Monitor Robot Condition and Environment?
Condition sensors do not always control robot motion directly. They can support overload detection, thermal protection, predictive maintenance, battery management, environmental monitoring and failure diagnosis.
| Measurement | Possible robot use |
|---|---|
| Current | Motor load estimation and fault detection |
| Temperature | Motor, drive, battery or gearbox protection |
| Vibration | Bearing, gearbox or structural-condition monitoring |
| Battery voltage/current | State and energy management |
| Pressure | Pneumatic, hydraulic or underwater systems |
| Humidity | Environmental qualification and enclosure monitoring |
| Magnetic field | Heading support and environmental detection |
What Is the Difference Between a Robot Sensor and a Safety Sensor?
A normal robot sensor provides information used by the robot application. A safety-related sensor is part of a defined safety-related system and must meet the required safety performance, diagnostic and fault-response requirements for that application. A sensor does not become safety-rated merely because it detects a person.
Examples: Safety laser scanner, light curtain, safety camera system, safety-rated position encoder, safety mat, interlocking switch and safety-rated proximity device.
Navigation LiDAR ≠ safety laser scanner. Standard camera ≠ safety-rated vision system. Ordinary encoder ≠ safety-related encoder.
Sensor selection alone does not establish robot or cell safety. The complete robot product and integrated application must be assessed under the applicable safety architecture and responsibility model.
contains additional requirements for diffuse-reflection electro-sensitive protective equipment. addresses the industrial robot itself, while addresses robot applications and cells.
Are Cameras Robot Sensors?
Yes. A camera is an optical sensor. However, a complete machine-vision system also includes optics, lighting, calibration, image acquisition, processing hardware and algorithms that turn image data into a measurement or decision.
See machine-vision cameras, optics, lighting and calibration for the complete visual-perception parameter universe.
How Do Robots Combine Multiple Sensors?
Sensor fusion combines measurements from multiple sources to estimate a robot state more reliably than one measurement can provide alone. The estimator uses a robot or motion model together with sensor measurements, timestamps and uncertainty information.
Examples: Wheel encoders + IMU, joint encoders + output encoders, LiDAR + IMU, vision + IMU, GNSS + IMU + wheel odometry, force sensor + motor current, tactile array + wrist force-torque sensor.
Sensor fusion architecture
- Sensor measurements
- Timestamp and frame alignment
- Filtering and outlier handling
- Uncertainty model
- State estimator
- Estimated pose, velocity or force state
- Controller or planner
Adding more sensors can degrade an estimate when timestamps, coordinate frames, covariance, biases or correlations are incorrect. Sensor fusion is not a substitute for sensor calibration.
What Are the Most Important Robot-Sensor Specifications?
Measurand and range
- Physical quantity, measurement axes, minimum and maximum range
- Overload range, dead zone and field of view
Measurement quality
- Accuracy, precision, repeatability, resolution and sensitivity
- Linearity, hysteresis, cross-axis sensitivity, noise, drift and bias
Dynamic performance
- Sampling rate, bandwidth, response time, latency, jitter
- Settling time and update consistency
Calibration
- Factory and field calibration, zeroing, temperature compensation
- Calibration interval, reference standard and calibration data access
Mechanical and environmental
- Mass, envelope, mounting, stiffness, overload protection
- Temperature, humidity, shock, vibration, ingress protection, cable flex life
Electrical, data and lifecycle
- Supply voltage, power, signal type, connector, communication protocol
- Data format, timestamp source, synchronization, diagnostics
- Expected life, recalibration, repair, firmware support and end-of-life policy
Accuracy vs Precision vs Resolution: What Is the Difference?
| Term | Meaning |
|---|---|
| Accuracy | Closeness of the reported value to the reference or true value |
| Precision | Closeness of repeated measurements to each other |
| Repeatability | Variation when measurement conditions are held defined and repeated |
| Resolution | Smallest represented or distinguishable measurement increment |
| Sensitivity | Change in output relative to change in input |
| Linearity | Deviation from the expected input-output relationship |
| Hysteresis | Different output for the same input depending on measurement direction or history |
| Drift | Change in output over time without the measured quantity changing |
| Bias | Systematic offset between measurement and reference |
| Noise | Random variation in the measured signal |
High resolution does not prove high accuracy. A sensor can report many digits while retaining systematic error, drift, noise or calibration uncertainty.
Sampling Rate vs Bandwidth vs Latency
| Parameter | Question answered |
|---|---|
| Sampling rate | How often is a measurement represented or transmitted? |
| Bandwidth | How quickly can the sensor meaningfully respond to changing input? |
| Response time | How long does the output take to react to a change? |
| Latency | How old is the physical measurement when the software receives or uses it? |
| Jitter | How much does measurement timing vary? |
| Synchronization error | How accurately do timestamps align across sensors? |
A high data rate does not establish high sensing bandwidth or low latency. The sensor may repeatedly transmit filtered, delayed or internally averaged data.
Required test questions: Is the timestamp generated at measurement or transmission? Does the data rate change under network load? What filtering is active? Is latency fixed or variable? Are multiple sensor channels sampled simultaneously? Can the sensor synchronize to an external clock?
Why Do Robot Sensors Need Calibration?
Calibration establishes the relationship between the sensor output and a known reference. Robot integration may also require geometric calibration: determining where the sensor is positioned and oriented relative to the robot, tool, joint or world coordinate system.
Measurement calibration
Relates signal output to a physical quantity.
Zero or bias calibration
Establishes the reference offset.
Intrinsic calibration
Models the sensor’s internal measurement behaviour.
Extrinsic calibration
Defines the sensor’s pose relative to another coordinate frame.
A factory calibration certificate does not prove that the sensor remains correct after installation. Mounting stress, cable forces, temperature, tool mass, magnetic fields, vibration or mechanical alignment may change the installed measurement.
specifies requirements for measurement-management systems intended to provide confidence in the validity and reliability of measurement results.
How Should a Robot Sensor Be Selected?
A robot sensor is not selected by measurement range alone. Its suitability depends on what must be measured, where the measurement is taken, how quickly it changes, how the data is calibrated and synchronized, and how measurement uncertainty affects the robot’s decisions.
1
Define the decision
- What robot decision or control action depends on this measurement?
2
Define the measurand
- Position, velocity, acceleration, force, torque, pressure, distance, temperature, current or contact
3
Define the measurement location
- Motor shaft, joint output, wrist, end effector, robot base, environment, workpiece or safety boundary
4
Define range and overload
- Normal range, peak range, minimum useful signal, overload event and survival load
5
Define required measurement quality
- Accuracy, repeatability, resolution, noise, drift, linearity and cross-axis error
6
Define dynamic behaviour
- Bandwidth, sampling rate, latency, jitter and response time
7
Define coordinate requirements
- Measurement axes, coordinate frame, orientation, cross-axis sensitivity and transformations
8
Define calibration requirements
- Factory calibration, installed calibration, zeroing, temperature compensation and calibration interval
9
Define mechanical integration
- Envelope, mass, mounting stiffness, overload protection, cable routing and replaceability
10
Define environmental conditions
- Temperature, humidity, dust, water, shock, vibration, EMC, lighting and target material
11
Define electrical and communication
- Voltage, power, analog or digital signal, EtherCAT, CAN, industrial Ethernet, serial or ROS support
12
Define synchronization
- Timestamp source, clock synchronization, trigger input and multi-sensor alignment
13
Define safety and failure behaviour
- Failure detection, diagnostic coverage, safe response, fallback sensing and redundancy
14
Validate in the real application
- Actual mounting, environment, target, robot vibration, software and communication load
What Interfaces Must Be Defined for Robot Sensors?
| Interface | Required information |
|---|---|
| Mechanical | Mounting, alignment, load, stiffness and tolerances |
| Electrical | Supply voltage, current, grounding and isolation |
| Signal | Voltage, current, frequency, pulse or digital data |
| Communication | Protocol, profile, address, baud rate and update behaviour |
| Coordinate | Axis definition, origin, orientation and units |
| Timing | Timestamp, clock, latency and synchronization |
| Software | Driver, API, message type, SDK and firmware |
| Calibration | Coefficients, matrix, zeroing and storage |
| Environmental | Temperature, IP rating, shock, vibration and EMC |
| Lifecycle | Repair, recalibration, firmware and replacement compatibility |
See robot-controller sensor interfaces for how measurements reach the motion-control system.
How Should Robot Sensors Be Tested?
Static measurement
- Range, accuracy, linearity, resolution, repeatability, hysteresis and cross-axis sensitivity
Dynamic performance
- Bandwidth, response time, sampling consistency, latency, jitter and phase delay
Drift and stability
- Zero drift, temperature drift, long-term drift, warm-up time and bias stability
Overload and durability
- Mechanical overload, shock, vibration, cable flex, connector life and ingress exposure
Calibration performance
- Calibration residual, recalibration repeatability, temperature compensation and reference traceability
Integration performance
- Mechanical installation, coordinate transformation, clock synchronization, communication loss and power interruption
Software and data integrity
- Units, timestamp, frame, missing data, diagnostic flags, firmware version and configuration retention
Application validation
- Real robot motion, real targets and surfaces, real environmental conditions and real controller and network load
Every sensor-performance result should identify sensor model and configuration, physical range, mounting, temperature, supply and interface, filtering, sampling and bandwidth settings, calibration status, reference equipment, test duration and uncertainty or tolerance.
contains environmental-test methods. defines enclosure ingress-protection classifications. These references support test planning but do not define the complete application-specific sensor validation.
Common Mistakes When Selecting Robot Sensors
| Mistake | Likely consequence |
|---|---|
| Selecting from range alone | Insufficient resolution or excessive noise |
| Treating resolution as accuracy | Overstated measurement quality |
| Ignoring bandwidth | Sensor cannot capture the real event |
| Comparing sample rate without latency | Control uses stale data |
| Ignoring drift | Measurement degrades over time |
| Ignoring cross-axis sensitivity | Loads or motion appear on the wrong axis |
| Using motor-side position as joint position | Gearbox error remains unmeasured |
| Ignoring calibration after installation | Mounting effects remain uncompensated |
| Adding sensors without synchronization | Fusion produces inconsistent states |
| Ignoring target material or surface | Distance sensor becomes unreliable |
| Selecting navigation LiDAR for safety | Inadequate safety architecture |
| Filtering data without documenting delay | Control response degrades |
| Ignoring cables and connectors | Intermittent field failures |
| Selecting prototype sensors without lifecycle support | Production redesign or obsolescence |
| Accepting supplier accuracy claims without conditions | Invalid supplier comparison |
What Evidence Should a Robot-Sensor Supplier Provide?
Product definition
- Measurand, axes, range, overload, mechanical drawing, mass, mounting, connector and pinout
Measurement evidence
- Accuracy, repeatability, resolution, linearity, hysteresis, noise, drift, cross-axis performance and test method
Dynamic and calibration evidence
- Sampling rate, bandwidth, latency, filtering, timestamp behaviour, synchronization
- Calibration method, reference equipment, certificate, coefficients and recalibration process
Environmental and integration evidence
- Temperature range, ingress rating, shock, vibration, EMC, cable-flex test
- Communication protocol, SDK or API, ROS driver, message format and diagnostic data
Manufacturing and lifecycle evidence
- Sensing-element source, signal-conditioning production, calibration process, end-of-line testing and traceability
- Warranty, repair, recalibration, replacement compatibility, firmware support and end-of-life notice
Confirmed through primary documentation
Supplier-reported
Observed or independently tested
Not confirmed
Not disclosed
Sourcing Robot Sensors in China
China has broad manufacturing capability across encoders, proximity sensors, load cells, force-torque sensors, inertial modules, LiDAR, environmental sensors, signal-conditioning electronics and integrated robot joints.
The central sourcing question is not whether the supplier can provide a digital sensor output. The buyer must establish who owns the sensing technology, which sensing elements and chips come from third parties, how calibration is performed, which test conditions support the published specifications and how software and lifecycle support are controlled.
Supplier evaluation should classify suppliers as sensor technology owners, sensing-element manufacturers, sensor-module manufacturers, signal-conditioning and interface suppliers, calibrated sensor-system suppliers, integrated robot-component manufacturers, private-label suppliers, distributors or trading companies.
Evaluate legal and operating identity, sensing-element ownership, mechanical sensing structure, analog-front-end and signal-processing ownership, calibration and reference equipment, temperature compensation, firmware ownership, communication-stack ownership, testing and traceability, critical imported components, calibration data retention, engineering-change notification, English technical documentation, overseas recalibration and repair, and product and firmware lifecycle.
A supplier may manufacture the housing and electronics while relying on a third-party sensing element. Another may own the sensing structure but use external calibration software or communication modules. The technology and responsibility map must therefore be established below the catalogue-product level.
What Should Be Included in a Robot-Sensor RFQ?
Robot and application context
- Robot category, application, sensor function and location
- Development stage, destination market, pilot quantity and expected annual volume
Measurement and dynamic requirement
- Physical quantity, axes, measurement range, normal and overload range
- Required accuracy, repeatability and resolution
- Bandwidth, sampling rate, maximum latency, permitted jitter and response time
Mechanical and environment
- Envelope, mass, mounting, axis orientation, structural stiffness
- Cable routing, connector, overload protection
- Temperature, humidity, dust, water, shock, vibration, EMC, lighting and target material
Electrical, software and calibration
- Supply voltage, power, signal type, communication protocol and data format
- Timestamp, driver, SDK and ROS message support
- Factory and installed calibration, zeroing, calibration interval and calibration-data access
Safety and lifecycle
- Safety-related use, diagnostics, fault indication, redundancy and required certification
- Expected life, recalibration, repair, spare parts, firmware support and end-of-life policy
Required evidence
- Datasheet, calibration report, accuracy and repeatability test
- Bandwidth and latency data, environmental tests, interface specification
- Manufacturing traceability and lifecycle policy
Frequently Asked Questions
What is a robot sensor?
A robot sensor measures a physical quantity and converts it into information the robot can use. Sensors may measure joint position, force, contact, distance, acceleration, temperature or other conditions. See the definition section and parent robot components guide.
What sensors do robots use?
Robots commonly use encoders, force-torque sensors, tactile sensors, proximity and range sensors, LiDAR, IMUs, current sensors, temperature sensors and cameras. The exact set depends on the robot and application. See the eight-category sensor taxonomy.
Why do robots need sensors?
Sensors allow robots to measure their own motion, detect objects and contact, estimate state, monitor condition and respond to changes instead of relying only on preprogrammed assumptions. See six engineering reasons.
What is the difference between internal and external robot sensors?
Internal sensors measure the robot’s state, such as joint position, current or temperature. External sensors measure the environment or interaction, such as distance, contact force or object position. See the comparison table.
How do robots know their joint position?
Robots commonly use encoders or resolvers on motors or joint outputs. Motor-side feedback measures motion before the gearbox; output-side feedback measures the actual joint after the transmission. See position and velocity sensors.
What is a six-axis force-torque sensor?
It measures force along three axes and torque around three axes, producing Fx, Fy, Fz, Tx, Ty and Tz measurements. See force and torque sensors.
What is a tactile sensor?
A tactile sensor measures local contact at a robot surface. It may detect pressure, shear, contact location, slip, texture or contact distribution. See tactile sensors and the distinction from wrist force-torque sensing.
What is the difference between LiDAR and an ultrasonic sensor?
LiDAR uses laser light and can provide higher angular detail and longer-range mapping. Ultrasonic sensing uses sound and is often simpler and lower cost, but has broader beams and different target-material effects. See the distance-sensor comparison and LiDAR section.
What does an IMU measure?
An IMU normally measures linear acceleration and angular velocity. It may also include magnetic-field sensing or an internal orientation estimate, but it does not directly measure global position. See inertial sensors and IMUs.
What is sensor fusion in robotics?
Sensor fusion combines multiple measurements and a system model to estimate a robot state. Examples include combining encoders, IMUs, LiDAR, cameras and GNSS for localization. See sensor fusion architecture.
Does higher resolution mean higher accuracy?
No. Resolution describes the smallest represented increment. Accuracy describes closeness to the reference value. A sensor may have high resolution but retain bias, drift or calibration error. See the measurement terminology table.
What is the difference between sampling rate and bandwidth?
Sampling rate describes how often data is represented. Bandwidth describes how quickly the sensor can meaningfully follow a changing physical input. A high sampling rate does not guarantee high bandwidth. See sampling rate vs bandwidth vs latency.
Why must robot sensors be calibrated?
Calibration relates sensor output to known physical references and may also establish the sensor’s position and orientation relative to the robot. See calibration and coordinate frames.
Can a normal LiDAR be used as a safety scanner?
Not automatically. Safety scanners must form part of a safety-related system and meet the relevant diagnostic, fault-response and safety-performance requirements. See safety-related sensors.
How do I select a robot sensor?
Start with the robot decision that depends on the measurement. Then define the measurand, location, range, accuracy, dynamic response, environment, calibration, interfaces and failure behaviour. See the 14-stage selection framework.
How do I evaluate a robot-sensor supplier?
Review the sensing technology, calibration system, test conditions, manufacturing controls, software interfaces, traceability and lifecycle support—not only the catalogue range and resolution. See the supplier evidence framework.
How do I source robot sensors from China?
Define the full sensor requirement first, then establish which sensing elements, electronics, calibration methods and software the supplier owns and which depend on third parties. See the China sourcing section and component sourcing service.
Need Help Sourcing Robot Sensors?
If you have defined the measurement requirement, integration constraints and calibration needs, Yana can help structure supplier research, sensor comparison and validation planning for robotics components in China.