The ROBART series of research prototypes has served the US Navy in developing the
component technologies needed in support of the US Army's Mobile
Detection Assessment Response System (MDARS) program. While ROBART I could only detect
a potential intruder, ROBART II could both detect and assess, thereby increasing
its sensitivity with a corresponding reduction in nuisance alarms. As the third-generation
prototype, ROBART III is specifically intended to demonstrate the feasibility of automated
response, using (for purposes of illustration only) a pneumatically powered six-barrel
Gatling-style weapon that fires simulated tranquilizer darts or rubber bullets.
More recently, ROBART III has also been employed
as a transition platform for the
Tech Transfer Program,
which harvests newly developed component technologies from various sources to increase the
functionality and autonomy of man-portable robots.
Jump to:
ROBART I (1980-1982)ROBART I
was Bart Everett's thesis project at the Naval Postgraduate School in Monterey, CA (Everett, 1982a; 1982b) and one of the very first behavior-based autonomous robots ever built. The navigation scheme provided a layered hierarchy of behaviors (see Table below) that looked ahead for a clear path (high-level), reactively avoided nearby obstacles (intermediate-level), and responded to actual impacts (low-level). A basic tenet of this strategy was the ability of certain high-level deliberative behaviors to influence or even disable the intermediate and low-level reactive behaviors, such as when docking with the recharging station, for example.
|
LEVEL |
BEHAVIOR |
RESULTING ACTION |
|
High |
Radar Dock |
Look ahead for encroaching obstacles Look for opening in forward hemisphere Home in on recharging station |
|
Intermediate |
Wander |
Seek clear path along new heading |
|
Low |
Proximity Reaction Impact Reaction |
Veer away from close proximity Veer away from physical
contact |
ROBART I's assigned function was to patrol a home environment, following either a random or set pattern from room to room, checking for unwanted conditions such as fire, smoke, intrusion, etc. The security application was chosen because it demonstrated performance of a useful function and did not require an end-effector or vision system, significantly reducing the required system complexity. Provision was made for locating and connecting with a free-standing recharging station when battery voltage began running low. Patrols were made at random intervals, with the majority of time spent immobile in a passive intrusion-detection mode to conserve power.
A Synertek SYM-1 single-board computer formed the heart of the onboard electronics. Speech synthesis (to allow the system to announce any unwanted conditions detected in the course of a random patrol) was implemented through National Semiconductor's Digitalker DT1050 synthesizer chip. Two sets of vocabulary instructions were stored on EPROMs for a total vocabulary of 280 words. A fixed vocabulary was chosen over an unlimited vocabulary created through use of phonemes in light of the greatly decreased demand on the onboard microprocessor in terms of execution time and memory space.
The software maintained the robot in one of two modes of operation: Alert Mode
or Passive Mode. In the Passive Mode, the majority of sensors were enabled,
but a good deal of the interface and drive control circuitry was powered down to conserve
the battery. The robot relied on optical motion detection, ultrasonic motion detection,
and hearing to detect an intruder, while at the same time monitoring for vibration
(earthquake), fire, smoke, toxic gas, and flooding (Everett, 1982a). Some of
these inputs were hard-wired to cause an alert (switch from Passive Mode to Alert
Mode), whereas others had to be evaluated first by software that could then trigger an
alert if required. Either mode could be in effect while recharging, and recharging could
be temporarily suspended if conditions so warranted.
Recharging was handled automatically. The 12-volt 20-amphour lead-acid battery gave about 6 hours of continuous service and then required 12 hours of charge. Roughly one hour of power was available to locate the charging station (by means of a visual homing beacon) after the battery monitor circuits detected a low condition. The homing beacon was activated by a coded signal sent out from an RF transmitter located atop the robot's head, and the recharging supply was activated only when a demand was sensed after connection. The robot could elect to seek out the recharging station before a low battery condition actually arose, such as between patrols.
The software employed in homing on the recharger and effecting a connection was able to deal with a multitude of problems that could arise to hinder the process. Provision was made to skirt around obstacles between the robot and the recharging station. If, as a result of a collision avoidance maneuver, the robot were oriented with respect to the charger so as to preclude a successful docking, the vehicle would back up and realign itself before continuing. The robot could also tell when a return from a forward-looking proximity detector was due to the presence of the recharging station, so the software would not try to steer the platform away.
A special near-infrared proximity sensor mounted on the head provided reliable
detection of diffuse wall surfaces for ranges out to 6 feet. This sensor could be
positioned at any angle up to 100 degrees either side of centerline by panning the head
and was extremely useful in locating open doors and clear paths for travel. Excellent
bearing information could be obtained, allowing this sensor to establish the location of
the edge of a doorway, for example, to within 1 inch of arc at a distance of 5 feet.
The hallway navigation scheme employed on ROBART I was based in part on the concept of beacon tracking. The recharging station optical beacon was suitably positioned in a known location to assist the robot in entering the hallway. Once in the hallway, the robot would move parallel to the walls in a reflexive fashion, guided by numerous near-infrared proximity sensors. General orientation in the hallway could be determined by knowing which direction afforded a view of the beacon. With a priori knowledge of where the rooms were situated with respect to this hallway, the robot could proceed in a semi-intelligent fashion to any given room, simply by counting off the correct number of open doorways on the appropriate side of the hall.
ROBART I was purposely intended to be a crude and simplistic demonstration of technical feasibility and was built on an extremely limited budget using oversimplified approaches. This philosophy assumed that if the concept could be successfully demonstrated under such primitive conditions of implementation, a reasonable extrapolation would show promise indeed for a more sophisticated second-generation version. (Bart Everett had actually started work on this follow-on prototype, ROBART II, just before leaving the Naval Postgraduate School in 1982. As his interests shifted more in this direction, ROBART I was loaned to the Naval Surface Weapons Center in White Oak, MD, entrusted to the watchful care of an MIT AI Lab co-op student by the name of Anita Flynn--now a famous pioneer in the field of microrobotics.) All work with ROBART I ended in 1985, when the prototype was shipped to Vancouver, BC, for display in the Design 2000 exhibit at EXPO '86. ROBART I is now on display alongside ROBART II in the museum section of our laboratory at SSC San Diego, CA.
For more information on ROBART I, see:
ROBART II (1982-1992)ROBART II became the center of focus for the next several years in Bart Everett's basement workshop in Springfield, VA. The system basically performed the same functions as its predecessor but employed a multiprocessor architecture to enable parallel real-time operations. Optimization of performance was addressed through significantly increased sensor capability, distributed processing, and precise vehicle motion control. Upon his transfer in 1986 to the Naval Ocean Systems Center (our predecessor), the prototype was made available to the Navy for use as a test bed in support of mobile robotics research. The initial development effort focused on two specific technology areas.
The first of these addressed the navigational concerns that were hindering successful implementation of a number of robotic applications requiring mobility (Gilbreath & Everett, 1988). Simply put, an autonomous vehicle must be able to determine its position and orientation in the workspace, plan a path to its intended destination, and then execute that path without running into any obstructions. Numerous proximity and ranging sensors were incorporated on the robot to support map generation, position estimation, collision avoidance, navigational planning, and terrain assessment, enabling successful traversal of congested environments with no human intervention.
The second thrust was aimed at producing a robust automated
security system exhibiting a high probability of detection with the ability to distinguish
between actual and nuisance alarms. ROBART II was therefore also equipped with a multitude
of intrusion and environmental sensors in support of its role as an intelligent sentry.
These sensors monitor both system and room temperature, relative humidity, barometric
pressure, ambient light and noise levels, toxic gas, smoke, and fire. Intrusion detection
is addressed through the use of infrared, optical, ultrasonic, microwave, and video motion
detection, as well as vibration monitoring and discriminatory hearing.
All high-level planning and assessment software runs on a desktop IBM-PC/AT computer connected to the robot via a 1200-baud Repco RF modem (Everett, et al., 1990). Robot position as well as sensor monitoring are represented graphically for the operator. The security assessment software package (Smurlo & Everett, 1993) displays time-stamped sensor status as well as environmental conditions, and can be overlaid on live video transmitted from a camera on-board the robot.
The scope of involvement was broadened in 1988 to include
enhancements to the world-modeling scheme to incorporate fixed-installation security
sensors (thereby allowing a mobile robot to operate in a secure area already protected by
installed motion sensors) and inventory monitoring capability (allowing the robot to
detect missing objects). In addition, a reflexive teleoperated control capability
was added in 1989 to free the operator from the lower-level concerns associated with
direct teleoperation. Speed of the vehicle and direction of motion are servo-controlled by
an onboard processor in response to local sensor inputs, but under the high-level
supervisory control of the remote operator (Laird & Everett, 1990). The
robot's rich array of collision-avoidance sensors, originally intended to provide an
envelope of protection during autonomous transit, were called into play during manual
operation as well to greatly minimize the possibility of operator error. The commanded
speed and direction of the platform was suitably altered as needed by the onboard
processors to keep the robot from running into obstructions.
In spite of having been built at home from hobbyist-grade components, ROBART II has
proven to be an amazingly reliable piece of equipment, with only four documented cases of
hardware failure since officially coming to life in early 1983. This record is somewhat
noteworthy, considering the workout given the system over its 20-year lifetime;
records indicate the robot performed in 53 live demonstrations for visiting students,
faculty, scientists, and government officials in 1987 alone. ROBART II was
continuously on line without a power interruption from 1988 to 2002, when someone accidentally
disconnected it's recharging station over a weekend.
For more information on ROBART II, see:
Like its predecessors, ROBART III is strictly a laboratory prototype, never intended for
fielding in the real-world: 1) it is not waterproof, 2) its mobility is constrained to planar
floor surfaces, so it can’t ascend or descend stairs, 3) it’s not defensively armored, 4) it’s
not rugged, and, 5) it cannot self right in the event it flips over.
Production versions of the physical security robot concept, which have been hardened and optimized
for fielding, are seen in the MDARS-Interior
and MDARS-Exterior platforms.
Production versions of tactical man-portable robots can be seen under the
Robotic Systems Pool section.
See also the
Technology Transfer Program section.
ROBART III is instead a concept-development platform optimally configured to support its intended
research and development role in a laboratory environment. Specific research thrusts include:
1) enhanced reflexive teleoperation, 2) automated target acquisition and tracking, 3) simultaneous
localization and mapping, 4) natural language understanding, and 5) augmented virtuality.
As a research platform, ROBART III is one of the most sophisticated mobile
robots in the world, and has been featured numerous times on the Learning, History,
and Discovery Channels. It was recently ranked number 16 in Wired Magazine's survey
of the 50 best robots ever (January, 2006).
RESEARCH THRUSTS Enhanced Reflexive Teleoperation
The basic issue being addressed here is the difficulty of controlling a
mobile robot equipped with a surveillance and/or targeting camera and also an
articulated weapon system. Experience
gained through extended use of conventional teleoperated devices of this type
has revealed considerable shortcomings from a man/machine interface point of
view. Simply put, if a remote operator
has to master simultaneous manipulation of three different joysticks (i.e., one
for drive and steering, another for camera pan and tilt, and yet a third for
weapons control), the chances of hitting a moving target are minimal.
In actuality, the single task of simply
driving a teleoperated platform using vehicle-based video feedback is no
trivial matter, and can be stressful and fatiguing even under very favorable
conditions.
The innovative thrust behind the ROBART III development is an extension of
ROBART II's reflexive-teleoperation (i.e, guarded motion) to incorporate
sensor-assisted weapon control into the integrated package. The approach
involves making two of the three controllable elements (i.e., drive control,
camera control, and weapon control) slaves to the third, so the human operator
only has to deal with one entity. For
example, the surveillance camera can be slaved to the weapon, so that the camera
looks wherever the operator points the gun.
If either the weapon pan-axis controller or the camera pan-axis
controller approach their respective limits of allowable travel, the robot's
mobility base automatically rotates in place in the proper direction to restore
the necessary range of motion. Alternatively, the weapon can be slaved to the
surveillance camera, and so forth. In all cases, final closed-loop control of
weapon pan-and-tilt can be provided by the video target-acquisition system.
Automated Target Acquisition and Tracking Simultaneous Localization and Mapping
For increased versatility as a prototype response vehicle, ROBART
III’s navigation strategy required modification to support fully autonomous
operation in previously unexplored interior structures (i.e., with no a priori
information or map). Starting in FY-03,
the navigation and collision avoidance schemes have been significantly enhanced
through technology transfer of improved algorithms developed under DARPA’s
Tactical Mobile Robot (TMR)
and Mobile Autonomous Robot Software (MARS)
programs (Pacis & Everett, 2004).
Under a Memorandum of Agreement, SSC San Diego has subsequently tasked the
Idaho National Laboratory (INL) to assist in
the coordinated development, evaluation, and transfer of robotics technology
that mutually benefits both Department of Defense and Department of Energy
missions.
As one of the key TMR contributors, SRI developed a mapping technique called
Consistent Pose Estimation (CPE) that efficiently incorporates new
laser scan information into a growing map.
Within this framework, SRI has addressed the challenging problem of loop closure: how to optimally register
laser information when the robot returns to an area previously explored.
With CPE,
it is possible to create high-resolution maps and repeatedly execute the
accurate path following necessary for high-level deliberative behavior.
Natural Language Understanding
SSC-San Diego is investigating a natural-language interface
that would allow a supervised autonomous robot to be given fairly unstructured
verbal direction, no different from the procedures used to instruct a human to
perform the same task. For example,
suppose the robot has penetrated an underground bunker and is streaming back
video that shows an open doorway in the center of the far wall of the room just
entered. A human monitoring this video
might converse with the robot as follows: “Find the doorway in
front of you.”
The robot would then analyze the current video, looking for predefined
scene attributes that suggest a door frame or opening, highlighting its choice
with a graphic overlay, then verbally respond,
“Request confirmation of selection.”
If the robot’s vision system locked onto the same doorway the
observer had intended, the human would acknowledge as follows: “Affirmative.”
If for some reason the robot selected the wrong door, or a set of scene
attributes that was in fact not a door at all, the human would respond
differently: “Negative, look to your left.” (Or right, as the case may
be.) The system would then shift focus
accordingly to the next set of scene attributes that looked like a doorway,
again ask for confirmation, and so forth.
Once the human and the robot were in sync, the human could issue
additional voice prompts to influence the robot’s further interaction with the
doorway. One example could be to zoom
in on and perhaps even illuminate for better assessment, or to enter the
doorway and continue searching on the other side.
HARDWARE DESCRIPTION Architecture Three 16-bit computers work together to give ROBART III its
advanced autonomous capabilities: 1)
the Torso Computer is the central computer system, responsible for gathering
and processing sensor data from the sonars, initiating speech output, and
controlling the integrated motion of the non-lethal weapon, and head; 2) the Vision Computer, located in the head,
is responsible for converting live video from the various cameras into a
sequence of digital images for processing;
and 3) the Drive Computer,
located in the Mobility Base, is responsible for controlling the drive motors,
and gathering sensor data from the Sick ladar, Sharp I/R rangerfinders, KVH fiber-optic
gyro, and the Microstrain compass. (In
addition, multiple 8-bit microcontrollers are employed for low-level sensor
processing and actuator control.) These
computers communicate over an Ethernet network and use the Player
protocol to share information. Mobility Base
Non-Lethal Weapon
The pneumatically-powered non-lethal weapon is for laboratory
demonstration purposes only. It supports the vision-based weapon-control
research without undue risk to personnel.
From a safety perspective, a local Ready/Standby
switch enables the air compressor and secondary accumulator charging, and a
local Arm/Safe switch physically interrupts power to the trigger solenoid
valve. There are parallel software
enables for both these same functions on the remote OCU.
Two separate control lines are employed for the
trigger solenoid, one active-high and the other active-low, to minimize chances
of inadvertent activation during initialization or in the event of a computer
reset. A three-way emergency override is also provided:
1) two local E-Stop buttons on the Mobility Base;
2) an RF kill pendent; and 3) a remote E-Stop button at the
control station.
A fiber-optic sensor on the gun is used to determine
load status for each barrel. (There is
also a remote accumulator dump valve that we don't typically use.)
For more information on ROBART III, see:
Videos of ROBART III: Please address questions/comments about this page to robo-web@spawar.navy.mil.
ROBART III (1992-)
Initial 360-degree motion detection is supported by a
ring of passive infrared sensors around the neck, an AM Sensors
microwave motion detector behind the face plate, and a Visual Stone omni-directional
camera mounted on the head. Fused
outputs from these sensors are used to cue a Canon high-resolution
pan-tilt-zoom (PTZ) camera in azimuth and elevation, which further assesses the
potential target. For static (i.e.,
motionless) targets, the PTZ protocol has been integrated with a
two-stage search-and-engage algorithm, wherein the
vision system first performs a wide-area scan for a pre-taught class of
objects, then cues the PTZ camera to zoom in and search for specific
“vulnerabilities” associated with that particular target.
The non-lethal weapon is automatically
trained accordingly with the aid of a bore-sighted targeting laser, and then
fired under operator supervision.
CPE is another method for performing simultaneous localization and mapping
(SLAM), based on original work by Lu and Milios (1997), who showed that information from a robot’s encoders and
laser sensors could be represented as a network of probabilistic constraints
linking the successive poses of the robot.
The encoders relate one robot pose to the next via dead-reckoning, and
the laser scans are matched to give further constraints between robot poses,
including constraints for when a robot returns to a previously-visited area (Gutman and Konolidge, 1999).
CPE provides an efficient means of generating a near-optimal
solution to the constraint network and yields high-quality metric maps.
Once a map has been made, it can be used to keep the moving robot localized.
The left and right drive wheels are driven by a pair of 12-volt electric wheelchair motors
identical to those used on ROBART II. System power is supplied by a 80-amphour
12-volt gel-cell battery which provides for several hours of continuous
operation between charges. Appropriate
hardware upgrades have recently been made to support the more sophisticated
navigation, collision avoidance, and mapping schemes, to include a MicroStrain
gyro-stabilized magnetic compass, KVH fiber-optic rate gyro, and a Sick
scanning laser rangefinder. Full-duplex
data communication with the PC-based host control station is accomplished via a
9600-baud Telesystems spread-spectrum RF link.
The non-lethal-response weapon chosen
for demonstration purposes is a pneumatically powered dart gun capable of firing a variety of 3/16-inch
diameter projectiles. The simulated
tranquilizer darts (20-gauge spring-steel wires terminated with 3/16-inch
plastic balls) illustrate a potential response application involving remote
firing of temporarily incapacitating rounds by military or law enforcement
personnel. A rotating-barrel
arrangement was incorporated to allow for multiple firings (six) with minimal
mechanical complexity. (The spinning-barrel mechanism also imparts a rather
sobering psychological message during system initialization.)
The darts are expelled at high velocity from their 12-inch barrels by a
release of compressed air from a pressurized accumulator at the rear of the gun
assembly. To minimize air loss, the solenoid-operated valve linking the gun
accumulator to the active barrel is opened under computer control for precisely
the amount of time required to expel the projectile. The gun accumulator is
monitored by a Micro Switch 242PC150G electronic pressure transducer,
and maintained at a constant pressure of 120 psi by a second solenoid valve
connected to a 150-psi air source. All six darts can thus be fired in rapid
succession (approximately 1.5 seconds) under highly repeatable launch
conditions to ensure accurate performance. A visible-red laser sight is
provided to facilitate manual as well as automatic targeting.
MPEG movie:
RobartCop faces 3 Coke cans, Western style (1.2MB).
MPEG movie of Robart
III shooting darts at a balloon (400K).
References:
Robotics at Space and Naval Warfare Systems Center, San Diego
SSC San Diego Adaptive Systems Branch