Unit 6: Robotic Sensor Applications and Use cases from Industry - Subjective Questions

Tactile sensing is the measurement of physical contact-related quantities such as force, pressure, vibration, temperature, and surface texture at the point of contact between a robot and an object.

Importance in robotics:

• It enables a robot to detect whether contact has occurred.
• It helps the robot regulate gripping force and avoid damaging delicate objects.
• It supports object recognition through surface texture, shape, and hardness information.
• It improves manipulation when visual information is insufficient or unavailable.
• It allows robots to detect slippage and adjust their grip in real time.
• It improves safety during human-robot interaction by detecting unexpected contact.

Typical tactile sensors include resistive, capacitive, piezoelectric, optical, and magnetic sensors. They are used in robotic hands, grippers, prosthetic devices, assembly systems, and collaborative robots.

Tactile sensors are integrated into the fingers, palms, or surfaces of robotic grippers to provide contact information during manipulation.

Working process:

• When the gripper approaches an object, the sensor detects the initial contact.
• Force measurements help the controller increase the gripping force gradually.
• Distributed tactile elements indicate the location and size of the contact area.
• Changes in tangential force or vibration can indicate that the object is slipping.
• The robot adjusts finger position and grip force using feedback from the sensor.

Applications:

• Handling fragile glassware, fruits, and electronic components.
• Sorting objects based on shape, hardness, or texture.
• Inserting parts into tight mechanical assemblies.
• Holding irregularly shaped objects securely.

This feedback-based approach is more reliable than controlling a gripper only through predetermined motor positions because it responds to actual contact conditions.

Soft robotics uses compliant materials such as elastomers, silicone, textiles, and flexible polymers to create robots that can deform safely around objects and people.

Role of sensors in soft robotics:

• Strain sensors measure bending, stretching, and deformation.
• Pressure sensors detect contact and internal pneumatic pressure.
• Tactile sensors measure external forces on soft surfaces.
• Temperature and proximity sensors support safe interaction.
• Embedded sensors provide feedback for closed-loop control.

Conventional rigid sensors may be unsuitable because they can restrict the robot's deformation, create stress concentrations, increase weight, and fail when repeatedly stretched or bent. Flexible sensors are designed to conform to changing shapes and maintain operation during deformation.

Sensor feedback allows a soft robot to estimate its body configuration, control its motion, regulate actuation pressure, and safely grasp delicate or irregular objects.

Rigid industrial manipulator:

• Usually uses encoders for accurate joint position measurement.
• Force-torque sensors are often mounted at the wrist.
• Sensors generally operate under predictable geometry.
• High repeatability and precise calibration are major requirements.
• The manipulator commonly performs fixed, programmed trajectories.

Soft robotic manipulator:

• Requires distributed strain and curvature sensing because its body continuously deforms.
• Needs flexible tactile sensors over large areas.
• Internal pressure sensors are important for pneumatic or hydraulic actuators.
• Sensor hysteresis, nonlinear deformation, and material fatigue must be considered.
• The robot often relies on feedback to estimate its shape and compensate for uncertainty.

Therefore, rigid robots emphasize joint accuracy and repeatability, whereas soft robots require body-state estimation, distributed sensing, adaptability, and compatibility with large deformation.

Robotic arc welding systems use several sensors to maintain the correct welding conditions and compensate for variations in workpiece geometry.

Important sensors include:

• Arc sensors: Detect changes in arc voltage or current during weaving and estimate the distance between the electrode and workpiece.
• Through-arc seam tracking sensors: Identify lateral deviations from the weld joint by comparing signals on different sides of the joint.
• Laser vision sensors: Locate the seam before welding and measure joint position, gap, and alignment.
• Touch sensors: Detect the workpiece or joint by making controlled electrical or mechanical contact.
• Temperature sensors: Monitor thermal conditions and help prevent overheating.
• Current and voltage sensors: Verify that the welding process remains within the specified operating range.

These sensors enable automatic seam tracking, adaptive control of torch position and speed, improved penetration, reduced defects, and consistent weld quality despite workpiece tolerances.

Through-arc seam tracking uses the welding arc itself as a sensing mechanism. The welding torch is moved in a controlled weaving motion across the joint.

Operating principle:

• The torch moves from one side of the joint to the other while maintaining the arc.
• Arc voltage or current changes as the torch-to-workpiece distance changes.
• The controller compares the measured signals at the left and right sides of the weave.
• A difference between the signals indicates that the torch is displaced from the joint center.
• The robot corrects its lateral position and may also adjust vertical distance.

Advantages:

• It operates during welding and does not always require a separate camera.
• It compensates for dimensional variation and joint misalignment.
• It improves bead placement and reduces incomplete fusion.

Limitations:

• It depends on stable arc conditions.
• It is less effective for very narrow joints or rapidly changing geometries.
• Calibration and suitable weaving parameters are necessary.

Contact-based methods:

• Use an electrically conductive wire, nozzle, probe, or welding electrode to detect the workpiece.
• Determine joint location through physical contact or electrical continuity.
• Are relatively simple and economical.
• May be affected by surface contamination, nonconductive coatings, or possible collision.
• Usually provide limited geometric information.

Non-contact methods:

• Use laser scanners, cameras, structured light, or optical triangulation.
• Measure joint position, gap, bead profile, and surface geometry without touching the workpiece.
• Can perform pre-weld inspection and real-time tracking.
• Provide richer geometric information and avoid probe wear.
• May be affected by arc brightness, smoke, spatter, reflective surfaces, and occlusion.

Contact sensing is useful for simple and robust joint detection, while non-contact sensing is preferred when detailed geometry and adaptive welding control are required.

Navigation sensors allow a mobile robot to estimate its position, perceive obstacles, build maps, and plan safe paths.

Indoor navigation:

• LiDAR or depth cameras measure walls, furniture, and obstacles.
• Wheel encoders estimate displacement using wheel rotation.
• Inertial measurement units estimate acceleration and angular motion.
• Cameras identify visual landmarks and support visual odometry.
• Ultrasonic sensors provide short-range obstacle detection.

Outdoor navigation:

• Global Navigation Satellite System receivers provide global position.
• LiDAR and cameras detect vehicles, pedestrians, terrain, and structures.
• Radar performs well in rain, fog, and dust and measures relative velocity.
• IMUs support motion estimation when satellite signals are blocked.
• Soil, slope, and vibration sensors may be useful for agricultural or field robots.

Sensor data is combined through sensor fusion to improve localization and obstacle avoidance. The navigation system then uses the estimated state to select and follow a safe path.

Sensor fusion combines measurements from multiple sensors to obtain a more accurate and reliable estimate than any single sensor can provide.

Example combination:

• Wheel encoders estimate short-term movement but accumulate drift.
• An IMU measures angular velocity and acceleration at high speed but also experiences bias and integration error.
• LiDAR or a camera identifies environmental features for correction.
• GNSS provides global position outdoors but may be unavailable indoors or obstructed near buildings.

A navigation filter, such as an extended Kalman filter or a particle filter, combines these measurements. The prediction step uses motion sensors, while the correction step uses landmarks or absolute position observations.

Benefits:

• Reduced localization drift.
• Better performance when one sensor is temporarily unavailable.
• Improved obstacle detection and map construction.
• Greater robustness under changes in lighting, weather, or terrain.

The selected sensors should have complementary error characteristics and overlapping operational coverage.

A physical sensor measures a physical quantity such as distance, force, temperature, pressure, acceleration, or light. A chemical sensor detects the presence, concentration, or properties of a chemical substance or gas.

Chemical sensor components:

• A sensing element that interacts selectively with the target substance.
• A transducer that converts the chemical interaction into an electrical or optical signal.
• Signal-conditioning and processing circuits.
• Calibration and compensation mechanisms for temperature and humidity effects.

Importance in robotics:

• Gas-detecting robots can locate leaks in industrial plants.
• Environmental robots can monitor air and water pollution.
• Agricultural robots can analyze soil nutrients and plant health.
• Healthcare robots can support breath or biochemical analysis.
• Inspection robots can identify hazardous or explosive atmospheres.

Chemical sensors are essential when a robot must perceive properties that cannot be determined from shape, motion, or visual appearance alone.

An autonomous inspection robot can carry gas and chemical sensors to monitor locations that may be dangerous for human workers.

Typical applications:

• Detection of methane, carbon monoxide, hydrogen sulfide, or volatile organic compounds.
• Identification of leaks in pipelines, storage tanks, and chemical processing equipment.
• Monitoring oxygen concentration in confined spaces.
• Detection of combustible gases before they reach dangerous levels.

System operation:

• The robot follows a planned route or investigates an abnormal reading.
• Chemical sensors sample the surrounding atmosphere continuously or at selected locations.
• The robot combines concentration measurements with its position estimate.
• A map of chemical concentration can be created to locate the source.
• When a threshold is exceeded, the system can stop, warn operators, or leave the area.

The robot should also use temperature, humidity, airflow, and navigation sensors because environmental conditions influence chemical measurements and source localization.

Wearable robotic devices include exoskeletons, prostheses, rehabilitation systems, smart garments, and assistive gloves. Sensors allow these devices to respond to the user's body and environment.

Major use cases:

• Motion monitoring: IMUs and joint encoders measure limb orientation and movement.
• Muscle-intent detection: Electromyography sensors detect electrical activity from muscles.
• Force measurement: Load cells and force sensors measure interaction between the user and device.
• Pressure monitoring: Pressure sensors detect foot loading, socket pressure, or contact between a glove and an object.
• Physiological monitoring: Heart rate, skin temperature, oxygen saturation, and respiration sensors track health status.
• Safety control: Sensors detect abnormal posture, excessive force, or loss of balance.
• Rehabilitation assessment: Sensor data measures range of motion, repetitions, gait symmetry, and recovery progress.

Sensor feedback enables personalized assistance instead of fixed mechanical operation.

An intention-aware exoskeleton estimates what movement the user is attempting and provides assistance at the appropriate time and level.

Sensor inputs:

• IMUs measure trunk and limb motion.
• Joint encoders measure the exoskeleton configuration.
• Force or torque sensors measure interaction forces between the user and the device.
• Electromyography sensors estimate muscle activation.
• Pressure sensors detect foot contact and the stance or swing phase of gait.

Control process:

• Sensor signals are filtered to remove noise and motion artifacts.
• Features such as joint angle, angular velocity, muscle activation, and foot pressure are extracted.
• The controller classifies the intended activity, such as walking, standing, or climbing stairs.
• Assistance is applied according to the detected phase and the user's required support.
• Feedback is continuously evaluated to prevent excessive force or unstable motion.

This approach improves comfort, energy efficiency, safety, and natural coordination between the wearer and the robot.

Sensors support data-driven agricultural operations by measuring crop, soil, weather, and equipment conditions.

Crop sensing:

• RGB and multispectral cameras identify plant rows, weeds, gaps, and visible disease symptoms.
• Hyperspectral sensors detect changes in plant chemistry and stress.
• Thermal cameras estimate water stress from canopy temperature.

Soil sensing:

• Moisture sensors support irrigation decisions.
• Electrical conductivity sensors estimate soil variation.
• Chemical sensors measure nutrients such as nitrogen and detect changes in soil chemistry.

Robot operation:

• LiDAR and cameras support row following and obstacle avoidance.
• Force sensors regulate contact during harvesting or weeding.
• Proximity sensors protect plants and machinery.
• Position sensors and GNSS support field coverage and data mapping.

These applications enable variable-rate irrigation, targeted spraying, robotic harvesting, weed removal, yield estimation, and early disease detection while reducing resource consumption.

A robotic weed-control system combines perception, localization, and contact sensing to remove weeds without damaging crops.

Sensor functions:

• RGB cameras detect color, shape, and plant boundaries.
• Multispectral or hyperspectral sensors distinguish vegetation using spectral reflectance.
• Depth cameras or LiDAR estimate plant height and three-dimensional position.
• Soil and proximity sensors help determine the position of the ground and nearby plants.
• Force sensors detect unexpected contact during mechanical removal.
• GNSS, wheel encoders, and IMUs maintain the robot's field position.

Processing sequence:

• The system captures and preprocesses images.
• Image or machine-learning algorithms identify crop and weed classes.
• The robot transforms plant coordinates from the camera frame into the tool frame.
• The actuator removes the weed mechanically or applies a localized treatment.
• Feedback verifies that the action was completed safely.

Lighting changes, occlusion, plant similarity, dust, and changing growth stages are important challenges.

Military robots use sensors for reconnaissance, surveillance, navigation, target identification, explosive ordnance disposal, logistics, and remote operation.

Common sensors:

• Daylight and infrared cameras for visual observation.
• Thermal imagers for night operation and heat-source detection.
• LiDAR and radar for mapping, obstacle detection, and movement tracking.
• Acoustic sensors for detecting sounds or locating events.
• Chemical and radiation sensors for hazardous-material detection.
• Inertial and satellite navigation sensors for localization.
• Force and tactile sensors for manipulation and bomb-disposal tasks.

Military sensing systems must often operate under poor visibility, dust, smoke, vibration, extreme temperatures, communication delays, and intentional interference. They therefore require rugged construction, redundancy, secure data transmission, low latency, and strong fault tolerance.

Civilian robots also require reliable sensing, but military systems typically face more severe environmental, security, and mission-critical constraints.

Search-and-rescue robots use sensors to locate people, understand dangerous surroundings, and operate where direct human access is unsafe.

Sensing applications:

• Thermal cameras detect body heat in darkness or through light obstruction.
• Microphones and acoustic arrays detect voices, tapping, or movement.
• LiDAR and depth cameras create maps of collapsed structures.
• Gas sensors detect toxic or combustible gases.
• Radiation sensors identify nuclear or radiological hazards.
• Inertial sensors and wheel encoders estimate the robot's motion.
• Tactile and force sensors help a manipulator handle debris or medical supplies.

Sensor data is combined to estimate victim locations and identify safe paths. Operators can use a map and sensor readings for teleoperation, while autonomous functions can support navigation and obstacle avoidance. The robot must provide reliable communication, stable localization, and safe interaction with unstable structures.

Healthcare robots use sensors for diagnosis support, rehabilitation, surgery, patient monitoring, assistance, and safe interaction with medical staff and patients.

Applications:

• Force-torque sensors regulate contact during rehabilitation and robotic surgery.
• Cameras and depth sensors track patient movement and posture.
• Tactile sensors help robotic hands handle medical instruments or objects.
• Vital-sign sensors measure heart rate, blood oxygen, temperature, blood pressure, and respiration.
• EMG sensors estimate muscle activity during prosthetic and rehabilitation control.
• Proximity sensors detect people and prevent collisions.
• Microphones and speech sensors support communication and patient assistance.
• Chemical and biochemical sensors can support analysis of breath, sweat, or bodily fluids.

Healthcare sensing systems must provide accuracy, patient comfort, electrical safety, privacy, ease of sterilization, and reliable operation. Sensor readings should be validated because incorrect measurements can directly affect clinical decisions.

Robotic surgical systems require precise sensing to control instruments inside or near the human body.

Position sensors:

• Encoders measure joint and instrument positions.
• They support accurate kinematic control and scaling of surgeon movements.
• They help prevent the instrument from exceeding safe motion limits.

Force and torque sensors:

• Measure interaction between the instrument and tissue.
• Detect excessive force that could cause tissue damage.
• Support force feedback or haptic feedback to the surgeon.
• Help maintain controlled suturing, grasping, and cutting forces.

Tactile sensors:

• Detect local contact and pressure distribution at the instrument tip.
• Help identify tissue stiffness, contact state, and grasp stability.
• Can support tissue classification when combined with other sensing methods.

Together, these sensors improve precision, safety, and repeatability. They must be highly accurate, sterilizable or protected, compact, and carefully calibrated.

Healthcare wearables:

• Measure human motion, muscle activity, pressure, and vital signs.
• Focus on personal safety, comfort, continuous monitoring, and individualized assistance.
• Often require low power consumption and direct skin contact.
• Must protect sensitive medical data and provide clinically meaningful measurements.

Agricultural robots:

• Measure crop condition, soil properties, weather, terrain, and machine interaction.
• Focus on field coverage, plant identification, resource optimization, and equipment robustness.
• Operate in dust, rain, sunlight, vibration, and changing outdoor conditions.
• Often combine cameras, multispectral sensors, chemical sensors, GNSS, and force sensors.

Both applications depend on calibration, sensor fusion, and reliable data interpretation. However, healthcare systems prioritize patient safety and physiological accuracy, while agricultural systems prioritize environmental robustness, scalability, and spatially distributed measurements.