Many applications for home robots can be proposed, but market-ready products have been slow to emerge. This may be due to insufficiently defined customer requirements or a failure to achieve an acceptable balance between affordability and functionality. This situation is beginning to change as processing power prices continue to decline. Furthermore, the widespread use of small, low-cost computer boards, such as the CircuitCo Beagle Board or Arduino boards, is enabling the global community of robot manufacturers to attract more talented individuals, thus creating conditions for producing suitable platforms that are commercially viable in the consumer market.
In fact, the first home autonomous robots have already appeared on the market: they come at a hefty price. Chores such as vacuuming floors and mowing lawns can now be delegated to machines capable of navigating throughout a room or lawn to clean or mow an entire area. Small robotic vacuum cleaners are available for around $200, while robotic lawnmowers are newer products on the market, priced at around $1200 or less.
Robot enthusiasts have shown the way forward.
Robot kits for builder enthusiasts are commercially available, such as the Parallax ActivityBot, and the guidance technologies needed to enable robots to perform household chores such as vacuuming or mowing have been well developed and understood. These have been honed in robot design competitions, such as the annual Nagoya International Miniature Robot Maze Competition (MAZE), which has been held since 1991.
AcTIvityBot is a basic educational robot that incorporates various technologies, including power electronics, processing, and sensing.
Algorithms for navigating line mazes can be based on one of two simple rules. For a basic maze, assuming no loops, success is guaranteed by either the right-hand rule or the left-hand rule. The right-hand rule always prioritizes a right turn over going straight or turning left. If there is no right-turn option, the right-hand rule will always go straight rather than turning left. Conversely, the left-hand rule prefers a left turn or going straight over a right turn.
The robot can use an array of infrared sensors to detect the path it needs to follow. These sensors illuminate the floor with infrared (or visible light) and monitor the reflections. Infrared reflective optical sensors, such as the Vishay TCND5000, may have been used. By cascading the sensors on the robot's chassis, left, right, or dead-point positions can be accurately detected using simple logic. For example, assume that light reflected from a white floor produces a high sensor output, while light from a dark line results in a low output. When the output changes, a T-junction in the line maze can be detected so that all sensors simultaneously indicate that they see black. Similarly, if the left sensor indicates black and the right sensor indicates white, a left or right turn can be detected, and vice versa. All sensors detecting white indicate a dead end. Assuming the chassis has independently driven wheels, the relative speeds of the left and right motors can be adjusted based on the detected patterns.
While maze competitions have helped accelerate the development of robot motion control, simple route tracking is not suitable for commercial equipment such as vacuum cleaners or lawnmowers. Different position sensing strategies are needed.
From challenges to housework
A complete robotic vacuum cleaner series uses optical interruptions to detect events, such as reaching the top of a wall or step, or detecting when the device has been picked up and should therefore be turned off. Optical sensors are mounted in a sensor strip at the front edge of the machine. These sensors can consist of discrete transmitters and detectors, or they can be integrated, such as the Omron EE-SX-3070 miniature photodetector, which includes both transmitter and photodetector elements, a built-in amplifier, and temperature compensation. The sensor strip has a molded bumper containing a beam blocker that moves between the transmitter and detector to interrupt the beam as the bumper deflects upon contact with a wall. In the case of the EE-SX-3070, its output is ON when the photodetector dims. As an alternative, the EE-SX-4070 provides an ON output when the detector is lit. The robot controller can use the generated signal as an indication that the edge of the room has been reached and can instruct the driven wheels to change direction.
Similar optical sensors or microswitches can be used to detect when the machine is picked up, activated when the wheels descend into its housing as the machine is lifted off the floor.
For stair detection, proximity to the floor is sensed by measuring optical reflection. Sensors such as the Vishay TCND5000 reflective optical sensor are suitable. The TCND5000 is a 6 mm x 4.3 mm surface-mount device with an infrared emitter mounted next to a photodetector in a package that excludes visible light. It can detect proximity ranging from 2 mm to 25 mm.
In robotic vacuum applications, this type of sensor is mounted downwards in a sensor strip at the edge of the machine, signaling that the stairs have been reached when the reflected signal suddenly weakens or is no longer detected.
More sophisticated non-contact wall sensing can be achieved using infrared or ultrasonic distance measurement. Infrared-based range sensors, such as the Sharp GP2Y0A51SK0F, generate an analog voltage proportional to the intensity of the infrared signal reflected onto the built-in detector. This sensor can measure distances from 2 cm to 15 cm, calculating distances using triangulation to minimize the impact of the surface reflectivity on measurement accuracy. The infrared emitter, detector, and signal processing circuitry are all integrated into a 27.0 mm × 10.8 mm × 12.0 mm package (Figure). The sensor offers fast start-up, producing the first stable output measurement within 5.0 ms. The typical output difference between the voltage corresponding to 15 cm and the voltage corresponding to 2 cm is 1.65 V.
Figure: The Sharp GP2Y0A51SK0F integrates the circuitry needed to generate an analog output voltage that is proportional to the measured distance.
Among other non-contact sensing technologies suitable for robot guidance, distance measurement using ultrasonic sensing has been proposed. Proximity sensing modules, such as the Parallax PING™ 28015 ultrasonic distance sensor, emit an ultrasonic "chirp" and generate an output pulse that terminates when an echo is received. The duration of this pulse represents the distance to the nearest reflecting surface (Figure). Placing multiple ultrasonic sensing modules on the front, rear, and sides of the housing allows the robot to detect its position relative to obstacles without damaging equipment or home decorations/furniture.
Figure: Distance measurement using an ultrasonic sensing module.
Just as a maze robot "learns" the maze it follows, a robotic vacuum cleaner can map the floor space it needs to clean. Through mapping, the robot can return to its docking station, for example, when sensors on the dust collector signal that it must be emptied, or when a battery charging status sensor indicates that it needs recharging. When ready to resume operation, the robot can return to its last working position to continue.
Let's go outside
Robotic lawnmowers must cope with a diverse set of operating conditions. There are often no convenient walls or other reflective surfaces to indicate the boundaries of the work area. Instead, a power line is buried or nailed around the lawn. Drive circuitry transmits signals along the boundary line; these signals can be radio frequency or audio signals, and sensors such as radio receivers, microphones, or inductive sensors mounted on the lawnmower detect when the device reaches the boundary. The lawnmower typically moves randomly within the wire mesh area to cut any grass longer than a pre-selected cut length. Safe cutting is crucial in such applications, including proximity sensors to avoid collisions and sensors to detect when the device is picked up or may be accidentally inverted.
Market Takeoff?
As affordable robots begin to appear in consumer applications, the next frontier of development may be airborne robotics. Parallax Inc., in addition to its semiconductor and component portfolio, offers educational kits, has a number of flying robot kits on the market, and is promoting an Unmanned Aerial System (UAS) education program. This aims to develop expertise across multiple disciplines, such as software design, motor control, electronic design including GPS, and the use of accelerometers and gyroscopes, as well as broader issues such as responsible flight. Potential applications may include emergency services rescue assistance, or consumer applications such as toys, new garden lighting, or aerial photography.
Walking, talking robotic personal assistants may remain the subject of science fiction or expensive research projects for some time to come. However, supported by mature, reliable, and affordable technologies such as high-performance processors and sensors, the era of home robots has begun.
