DC Velocity - November 2016

magnetic field, leading the vehicle to follow the tape.

A variation on this theme is a magnetic grid, which uses magnets affixed to or embedded in the floor in a grid pattern. An onboard sensor detects the magnets, and the reference points are stored in the AGV’s memory as X and Y coordinates. A gyroscope on the vehicle measures and maintains direction, and a wheel encoder calculates the distance traveled. In a magnetic grid, the guide paths can easily be changed.

Still another option is inertial navigation, where transponders are embedded in the floor. An onboard gyroscope detects slight directional changes and corrects the vehicle’s travel path to keep it on course. Providers such as Daifuku’s Jervis B. Webb division note that inertial guidance vehicles can operate in almost any environment, including tight aisles and extreme temperatures.

NEWER KIDS ON THE BLOCK

More recently developed guidance technologies rely on var-
ious ways of measuring distances, mapping, storing data,
and decision making for navigation. All provide a degree
of flexibility that earlier technologies couldn’t offer—
probably the biggest reason for the inroads AGVs are now
making in warehouses and DCs. They all make it easy and
fast to reprogram routes, require no (or, in the case of laser
guidance, minimal) additional infrastructure, and can nav-
igate on their own around obstacles.

Laser-guided vehicles map and store the facility layout in the vehicle’s computer. A laser transmitter/receiver mounted on the vehicle detects reflective strips located at fixed reference points and measures both its distance and angle relative to the reflectors. By triangulating two reference points, the AGV can determine and update its location. AGV maker JBT Corp., for example, says its patented laser-guidance technology uses an eye-safe laser scanner that “strobes” the operating area and updates its position several times per second, resulting in highly accurate positioning. Transbotics, another AGV developer, touts laser guidance for its accuracy, reliability, security, dynamic traffic management, and short installation times.

Natural-feature guidance is a relative newcomer to the AGV scene. AGVs equipped with this type of technology record and store reference images as a map of the operating area. They then navigate by calculating their position relative to existing features—walls, racks, I-beams, doorways, stacks of pallets, and so forth—following the most efficient path, just as a human being would when walking through the facility. A major advantage of this technology is that it requires no markers, transponders, or reflectors. In addition, guide paths can easily be changed by retraining the AGV or by drawing a new route on the map. Sweden’s Kollmorgen was one of the first to develop this capability, and others have followed. AutoGuide, for example, is about to introduce a low-profile AGV that measures the locations of natural features to use as reference points as it moves along its route, says Sarah Carlson, vice president of marketing and business development.

In somewhat similar fashion, the Otto Motors division of Clearpath Robotics uses simultaneous localization and mapping (SLAM) technology for its self-driving material handling vehicles—the same underlying technology used in self-driving highway vehicles, says Simon Drexler, director of industrial solutions. Otto uses laser-based “lidar” (from “light” and “radar”) scanning to gather data and construct a highly detailed map of the facility floor. Once it has the reference map, it can navigate any route without a defined path or line. The vehicle is intelligent enough to plan and follow its own route, Drexler says. Once the reference map is in place, users can drag and drop location pins on the map to instruct the vehicle where to stop for pickups and dropoffs.

Vehicles that use vision-based navigation come closest to processing visual information the way a human being does. AGVs built by Seegrid, which pioneered this technology, use five pairs of stereo cameras to record the surrounding environment as an operator “trains” them by walking them through their route. The cameras take two images simultaneously, achieving binocular vision with

54 DC VELOCITY NOVEMBER 2016 www.dcvelocity.com
m
at
er
i
a
lh
a
n
d
l
i
ng
u
p
d
at
e
AU
TOMA
T
ED
GUID
ED
V
EHI
C
L
E
S

Who makes automated
guided vehicles?

Just a few years ago, only a handful of companies were designing, making, and selling automated guided vehicles (AGVs). Today, there are quite a number of vendors and types of automated vehicles, including load carriers, load lifters, driverless forklifts, tuggers, low-profile carriers, and automated carts.

Interested in checking them out? This is by no means
a comprehensive list, but the following are some of the
AGV providers we’ve run across:
AutoGuide ( www.autoguideagvs.com)