A photoelectric proximity sensor uses a light-beam generator, a photodetector, a special amplifier, and a microprocessor. The light beam reflects from an object and is picked up by the photodetector. The light beam is modulated at a specific frequency, and the detector has a frequency-sensitive amplifier that responds only to light modulated at that frequency. This prevents false imaging that might otherwise be caused by lamps or sunlight.

An acoustic proximity sensor works on the same principle as sonar. A pulsed signal, having a frequency somewhat above the range of human hearing, is generated by an oscillator. This signal is fed to a transducer that emits ultrasound pulses at various frequencies in a coded sequence. These pulses reflect from nearby objects and are returned to another transducer, which converts the ultrasound back into high-frequency pulses. The return pulses are amplified and sent to the controller. The delay between the transmitted and received pulses is timed, and this will give an indication of the distance to the obstruction. The pulse coding prevents errors that might otherwise occur because of confusion between adjacent pulses.

A capacitive proximity sensor uses a radio-frequency (RF) oscillator, a frequency detector, and a metal plate connected into the oscillator circuit. The oscillator is designed so that a change in the capacitance of the plate, with respect to the environment, causes the frequency to change. This change is sensed by the frequency detector, which sends a signal to the apparatus that controls the robot. In this way, a robot can avoid bumping into things. Objects that conduct electricity to some extent, such as house wiring, animals, cars, or refrigerators, are sensed more easily by capacitive transducers than are things that do not conduct, like wood-frame beds and dry masonry walls.