Sensors: A Robot's Inputs
A robot that cannot sense its environment is like a person trying to navigate a dark room with no sight, no touch, and no hearing. Without information from the outside world, the controller has nothing to act on — it is making decisions blind. Sensors are the devices that bridge the physical world and the robot's digital brain. They detect a physical quantity — light, distance, pressure, temperature, rotation — and convert it into an electrical signal that the controller can read as a number.
The technical name for any device that converts one form of energy into another is a transducer. A microphone is a transducer: it converts sound waves into electrical voltage. Every sensor in a robot is a transducer — it converts a physical quantity (light intensity, distance, force) into the electrical signal the controller understands.
Common Sensor Types
Proximity and distance sensors tell the robot how far away objects are. Ultrasonic sensors emit a pulse of high-frequency sound and measure how long it takes to echo back — longer echo means farther away. Infrared sensors bounce an invisible beam of light off nearby surfaces. LiDAR (Light Detection and Ranging) fires rapid pulses of laser light and builds a detailed 3D map of the surrounding space. Self-driving cars use LiDAR to see everything within about 100 meters at high resolution. Camera sensors capture images as grids of pixels. A camera does not tell the robot what it is seeing — that interpretation requires software, often powered by machine learning. But cameras give robots an enormous amount of raw information: color, shape, texture, and relative position of objects. Touch and force sensors detect physical contact. A simple touch sensor is just a switch: pressed or not. A more sophisticated force sensor measures how hard a surface is being pushed. Surgical robots use ultra-precise force sensors to detect the resistance of different tissue types — essential when the difference between enough force and too much could injure a patient. Inertial measurement units (IMUs) contain accelerometers and gyroscopes. An accelerometer measures how quickly the robot is speeding up or slowing down in each direction. A gyroscope measures the robot's rotational rate — how fast it is spinning or tilting. Together, these sensors tell the robot how it is moving even when it has no view of the outside world. Drones rely heavily on IMUs to stay level. Temperature sensors detect heat. Environment-monitoring robots, food-safety inspection robots, and firefighting robots all use temperature sensors to detect heat sources and avoid or approach them accordingly. Magnetic encoders are attached to motors and measure how far and how fast the motor shaft has rotated. They give the controller precise feedback on whether the motors are doing what they were told — a key part of making a robot move exact distances.
Real robots rarely rely on just one sensor for a critical task. They combine data from multiple sensors — a technique called sensor fusion. A self-driving car combines cameras, LiDAR, and radar because each has different strengths and weaknesses. Cameras see color but struggle in rain. LiDAR is precise but expensive. Radar works in bad weather but has low resolution. Together they cover each other's gaps.
Match each sensor to the physical quantity it primarily measures.
Terms
Definitions
Drag terms onto their definitions, or click a term then click a definition to match.
Choosing the Right Sensor
Every sensor has tradeoffs. Range, precision, cost, power consumption, and environmental sensitivity all matter. An ultrasonic sensor costs a few dollars and works well for a robot avoiding walls in a hallway, but it cannot distinguish a cardboard box from a person. A LiDAR unit provides detailed 3D maps but costs hundreds or thousands of dollars and needs significant processing power. Choosing the wrong sensor for the job means the robot either has information it does not need or lacks information it does need — and both scenarios lead to poor performance.
Sensors are not perfectly accurate. Every sensor measurement contains some amount of noise — random variation caused by electrical interference, temperature changes, or physical imperfections. Engineers must write software that filters out noise before trusting sensor readings. A robot that acts on every noisy reading without filtering will behave erratically.
A robot in a competition must navigate a maze and stop exactly 5 cm from each wall. Which sensor combination would give it the most useful data for this task?
What does a transducer do, and why are sensors described as transducers?
Flashcards — click each card to reveal the answer
Your Body as a Sensor Array
- Step 1: List every sensor your human body has — sight, hearing, touch, smell, taste, balance, proprioception (knowing where your limbs are), temperature, and pain.
- Step 2: For each, name the physical quantity it detects and describe how the signal reaches your brain.
- Step 3: For each human sense, name the closest robot sensor equivalent (or write 'no equivalent yet' if none exists).
- Step 4: Pick one task — catching a ball, tying a shoelace, recognizing a friend across a room. List every sensor involved and explain why removing any one of them would make the task harder.
- Step 5: Identify two tasks where robots already outperform humans at sensing, and two tasks where human sensing still beats the best robot sensors available today.