Integrating Odor Sensing in Robotics

Introduction

Among the challenges facing robotics researchers is developing intelligent autonomous robot that can take decision by his own. This kind of robot should rely on complex sensors to gather the necessary information about the environment it navigates and should also use intelligent algorithms that enable the robot to take the necessary actions when needed. The robot consists mainly of many parts: the sensors, the controller, effectors and actuators. Sensors are used to provide the robot with necessary information about the environment in which the robot is navigating and discovering. Things that can be sensed includes light and sound (presence, color, intensity, content (mod), direction), sound (presence, frequency, intensity, content (mod), direction), heat (temperature, wavelength, magnitude, direction), chemicals (presence, concentration, identity, etc.), object proximity (presence/absence, distance, bearing, color, etc.), physical orientation/attitude/position, magnitude, pitch, roll, yaw, coordinates, etc. Commonly used sensors includes ultrasonic sensor, infrared sensor, touch sensor, sonic sensor, laser sensor, photoelectric light sensors, magnetic sensor and recently chemical sensor. Each one of these sensor measure different environmental phenomena and provide the robot with different information. Sensors that mimic the animal world (particularly the human senses) are finding their use in robotics like sensors for taste and smell (chemical sensors). Other part of robot is the controller which receive signals from sensors, analyze these signals and take the appropriate action. The controller can be based on deliberative or reactive model. The deliberative model is representation dependent, need complex sensors, uses high level intelligence and therefore has slower response and variable latency. On the other side, the reactive model is representation free, uses low level intelligence and therefore it performs simple computation and has real time response. The effectors and actuators terms are often used interchangeably to mean: “whatever makes the robot take an action” but they aren’t the same thing. An effector is any device that interact with the environment and is controlled by the robot controller. It can range from legs and wheels to arms and fingers that is used to grip objects. An actuator is the actual mechanism that enables the effector to execute an action. It can include electric motors (AC or DC motor, servo motor and stepper motor), hydraulic cylinders, and pneumatic cylinders etc. Robot can be autonomous or tele-operated. The autonomous robot can navigate and move without human intervention while tele-operated robot is controlled by human operator. Tele-operated operator can be wired controlled or wireless controlled. Wireless controlled can be achieved using infrared, radio frequency, bluetooth, WiFi or cellular/ GPRS signal. Robot also can work on land, air or underwater. Robot that works on land is called unmanned ground vehicle (UGV). Robot that works on air is called Unmanned Aerial Vehicle (UAV) or drone and finally robot that works underwater is called Unmanned Underwater Vehicle (UUV).