http://www4.army.mil/news/article.php?story=8606
Army testing unmanned Stryker convoys
By Larry Edmond
February 21, 2006
FORT GORDON, Ga. (Army News Service, Feb. 21, 2006) -- Engineers conducting show-and-tell with a 20-ton robot on the last day of two weeks of trials on Fort Gordon were cautiously optimistic.
Karl Murphy, a software engineer from Robotic Research, said there was a new principle of “Murphy’s Law” at work on the test field Feb. 10.
"One of my professors reminded us that we have 'sight-ons' present whenever an experiment is being viewed,” Murphy said. “The more 'sight-ons' you have, the greater is the potential for something to go wrong."
Tongue in cheek, he continued explaining that sight-on fields increase with the rank and reach of individuals viewing a test. With national, regional and local media rolling cameras, the “sight-on” field was very high that Friday.
The demonstration at Fort Gordon was a part of a much larger program of tests being conducted by the Robotics Technology Integration Team from the U.S. Army Tank Automotive Research and Development Engineer Center, General Dynamics Corporation and its sub-contractors.
Jeff Jaczkowski, TARDEC electrical engineer and manager for this Robotic Follower Advanced Technology Demonstrator project, explained what the testing was about and why Fort Gordon was chosen.
Pointing to the two Stryker Infantry Carrier vehicles idling behind him at their base camp on Range 37, Jaczkowski said these vehicles are part of a larger program set to bring vehicle electronics-vetronics technology integration and robotic systems to the force.
The system in testing at Fort Gordon is the robotic follower program. This program seeks to develop robots that can conduct convoy operations. One of the vehicles is called the CAT - short for crew integration and automation test bed. It serves as the manned leader vehicle.
The other vehicle serves as an unmanned follower in a convoy.
"We have done a circuit of testing that started in 2003,” Jaczkowski said. “We have done a number of different environments, including Fort Bliss, Texas, that has a primarily sand/desert environment. We did Fort Knox, Ky., where there is more cross-country terrain. There was Fort Indiantown Gap and Letterkenny Army Depot in Pennsylvania.
"We are down here at Fort Gordon for the environment in the forested-type setting. We are focusing on road and long-haul convoy missions."
Jaczkowski said Fort Gordon provides an ideal setting with a 10-kilometer loop that has a three-kilometer stretch of dirt road and the rest is paved. The long stretches of isolated roads allow the teams to put their test vehicles through a series of high-speed tests.
"Yesterday we ran a 100-mile test where the lead vehicle was being driven manually and the robot was following,” Jaczkowski said. “We did this successfully where the average speed was about 22 miles per hour. You may think that 22 miles per hour is not that fast when operational convoys are going 60 to 70 miles per hour. But you have to take into account that we did 68 right turns.
“You don't take right turns at 50 miles per hour, especially with a 20-ton robot."
On straight stretches, the vehicles routinely speed along at more than 40 mph, Jaczkowski said.
Pointing to the bristling array of sensors on the vehicles, Jaczkowski said these vehicles incorporate second-generation ladar-laser radar, forward-looking infrared sensors, and advanced computers to handle autonomous navigation.
For the autonomous follower, engineers are going beyond Global Positioning Systems to link terrain data from the lead vehicle back to the follower vehicle to augment data the follower vehicle gathers from on-board sensors.
"We have a major emphasis to create systems that can operate without GPS. We know that electronic interference can easily jam GPS in a battle zone.
"The idea is to pass electronic bread crumbs from the manned lead vehicle back to the autonomous follower vehicle, and provide high-level proofing of the follower's path so the follower avoids areas that might impede or confuse its autonomous navigation system, while requiring only a minimum of human intervention and control from the lead vehicle," Jaczkowski said.
Jaczkowski characterized all the testing so far as outstanding and gave high marks to the Fort Gordon Battle Lab and range control. "
Jaczkowski is quick to point out that the testing that he is conducting is not about the relatively new Stryker vehicle.
"The Stryker is a fielded system, but the robotic convoy technology sensors that we have on these units are what we are putting through the research and development stage."
The demonstration conducted before the media Feb. 10 showed how adept the robot is in making decisions. The lead vehicle was manually driven along the road through an area where a gate wa set, with the robot vehicle following about 100 meters behind.
After the lead vehicle passed, the engineers were planning to pull a cord releasing a gate to block the path. The follower vehicle should be able to detect the gate and plot a path around it before continuing, Jaczkowski said.
True to Murphy's Law and the sight-on rule, the chord attached to the spring release broke as engineer Karl Murphy tugged it to release the gate. Undaunted, Murphy reached down and tripped the release. The gate swung into the path of the following robot.
With only a few seconds to assess, the robot slowed, and veered around the gate. It then continued on its path, following the lead vehicle.
It was a slight glitch and only served to more graphically demonstrate how perceptive the robot is, Jaczkowski said.
In the future, Jaczkowski speculates the current efforts will lead to manned and unmanned convoys.
"There are two avenues that the Army is pursuing. The near-term objective is to automate the function of driving in a convoy vehicle."
Soldiers will remain in the vehicles for now, but by placing a vehicle on auto-pilot, the driver will be able to perform other duties or rest.
The long-term objective is to create dedicated unmanned ground vehicles.
In the tests being conducted on Fort Gordon, the lead vehicle develops a path along a route that it transmits to the follower vehicle that can follow the path immediately or weeks later.
"This is the beginning of going from point A to point B autonomously," Jaczowski said.
He said the payoff will be in saving lives from such routine missions as resupplying forces in environments like Iraq where roadside bombs wreak havoc.
The group will continue testing Feb. 24 to March 10, Jaczowski said.
(Larry Edmond serves as a staff writer for the Fort Gordon Signal newspaper.)
本城市首頁
看回應文章
等級：8
