AI in the News. 26/9/2006.
“As the U.S. Army transforms into a lighter, more lethal force, the need for small mobile weapons systems (SMWS) becomes more crucial. Unmanned aerial vehicles (UAVs) have already shown great advantage as an extension of the soldier for RSTA (reconnaissance, surveillance and target acquisition) missions, and SMWS are becoming available to provide a critical multiplier of the firepower in a transformed force.”
“TALON robots can be configured with M240 or M249 machine guns or Barrett 50-caliber rifles for armed reconnaissance missions. A prototype system was delivered to the 3/2 Stryker brigade for evaluation, and successful testing was performed by the brigade in Kuwait in December 2003.
Additional prototypes have been manufactured and are currently undergoing system safety certification by the U.S. Army.
Alternative weapons, including 40 mm grenade launchers and anti-tank rocket launchers, continue to be evaluated by the U.S. Army.”
TALON brochure from Foster Miller
The new, armed version of the robot, TALON™ - SWORDS, was recognized by Time magazine as one of the “most amazing inventions of 2004.” It can be equipped with several different weapons and is then operated remotely by the soldier.
Foster-Miller's Sword is a variant of Talon in which the manipulator arm has been replaced by a rotating machine-gun carrier.
"It's for urban combat and perimeter security and it's fully controlled by the soldier," Quinn says. Touted uses include checking out a potential ambush.
Apart from a planned autonomous "return home" function, the Sword prototype does not operate autonomously.
Bob Quinn, general manager at Foster-Miller of Waltham, Massachusetts, whose machine-gun-equipped robot, Sword, was certified safe for use by the US forces in June (2006), has said that robot infantry may soon became a reality …
"Sometime in the coming months, chances are that we'll be seeing TV reports that an armed remote-controlled robot has been used in anger for the first time. They will appear when they appear. I can't talk about when that may be"
PS: The suspect dropped his weapon but a fault in the robot's software caused it to open fire anyway.
The DoD wants to engineer mobile robots to "understand cooperative and uncooperative" people, and inform their operator if they seem a threat.
It hopes to do this using artificial intelligence software fed with data from a "remote physiological stress monitoring" system, and by using speech, face and gesture recognition.
From this it would draw inferences about the threat that person poses.
(Small Business Innovation Research, Small Business Technology Transfer)
The Department of Defense (DoD) SBIR and STTR programs fund a billion dollars each year in early-stage R&D projects at small technology companies -- projects that serve a DoD need and have commercial applications.
The SBIR Program provides up to $850,000 in early-stage R&D funding directly to small technology companies (or individual entrepreneurs who form a company).
The STTR Program provides up to $850,000 in early-stage R&D funding directly to small companies working cooperatively with researchers at universities and other research institutions.
Small companies retain the intellectual property rights to technologies they develop under these programs.
Funding is awarded competitively, but the process is streamlined and user-friendly.
Deputy Director Of Defense Research & Engineering
Deputy Under Secretary Of Defense (Science & Technology)
Small Business Innovation Research (SBIR)
FY2006.3 Program Description
Announced August 1, 2006
Proposal Deadline October 13, 2006
The DoD is currently investing in a variety of unmanned systems designed for units at the battalion level and below.
In many of these cases, there will be significant limitations on the manning available for these systems in terms of both numbers and skill types.
Increasing the level of automation can have a significant impact on
reducing manning requirements.
Despite many advances in autonomous control technologies, mission management often still requires a human's cognitive skills, judgment, decision-making, and tactical understanding.
Further, future unmanned systems missions may require frequent user interactions with the autonomous agents to coordinate autonomous planning and execution with those of manned platforms or units in a dynamic battlespace.
This is particularly relevant to the rapidly changing environment of the global war on terrorism with the challenges of littoral and urban operations, the threat of chemical and biological weapons, the difficulty in differentiating enemies from neutral civilians, and the need for better force protection.
Because of the major threat that Improvised Explosive Devices (IEDs) have posed to troops in Iraq and the importance that urban and littoral operations have assumed in recent conflicts, mobile robots are being rushed into service in large numbers.
However, many of these systems have significant manning requirements in terms of both the numbers and skills of operators required to operate these systems effectively.
This is particularly challenging in a dynamic battlespace environment.
The need for improved human control and collaboration with these robots is acute.
One issue that limits the use of robotic and autonomous systems
in urban environments is their inability to recognize and interact with persons who may be either non-combatants or threats.
For example, soldiers manning checkpoints are exposed to vehicle-borne
IEDs, but replacing them with robots requires advanced capabilities in vision and communication between robots and human subjects.
Another acute need is improved situation awareness regarding threats from dismounted persons around vehicles, both manned and unmanned, and mobile robots in urban environments.
Recent research on spatial aware cognitive models embedded on mobile robots has shown promise for significantly enhancing human robot interactions by providing a framework for robot understanding of human goals, plans, activities and communications.
Research in machine vision is developing new capabilities in human activity and gesture recognition.
• Human Robotic Interaction (HRI) in autonomous operations.
• Computational cognitive models as reasoning agents; affective computing to support social regulation.
• Multi-modal and mixed-initiative interaction; human-guided learning; dynamic autonomy.
• Team collaboration with autonomous vehicle team members; HRI for heterogeneous teams.
• Joint human-robot manipulation and mobility.
• Advanced embedded vision supporting HRI, including tracking, gesture and activity recognition.
• Omnidirectional EO/IR and acoustic sensors and software that provide 360 degree awareness around a vehicle, particularly in enclosed urban spaces. Ability to detection and interaction with humans that approach from any direction. Ability to isolate conversation for automated translators and focus attention on threats.
• Human understanding or mental models of intelligent agents that enhance human robot interaction. Integrated computational theories of human robot interaction.
• Natural language interaction for high-level human control of robotic agents. Robotic speech and face avatar communication capabilities on UGVs/USVs.
OSD06-UM1 Littoral Navigation Autonomy for Unmanned Surface Vehicle (Navy)
OSD06-UM2 Cooperative Tracking of Elusive Dismounts by Human Assisted UAV-UGV (Air Force)
OSD06-UM3 Human-Robot Manipulation for Complex Operations (Navy)
OSD06-UM4 Command and Control of small robotics assets (Army)
OSD06-UM5 Peer-to-Peer Embedded Human Robot Interaction (Navy)
OSD06-UM6 Collaborative and Shared Control of Unmanned Vehicle Systems (Navy)
OSD06-UM7 Affect-Based Computing and Cognitive Models for Unmanned Vehicle Systems (Navy)
OSD06-UM8 UAV – Combat Medic Collaboration for Resupply & Evacuation (Army)
TECHNOLOGY AREAS: Ground/Sea Vehicles, Weapons
OBJECTIVE: Develop methodologies and technologies to enable mobile robots to communicate with and understand the actions of dismounted persons encountered and to collaborate with single operators as peers.
Develop a concept for peer-to-peer embedded human-robot interaction that
exploits sensing of human actions and communications including some combination of:
• machine vision techniques for human activity and gesture recognition
• omnidirectional detection and tracking of people in relation to a mobile robot
• head finding, head pose recognition, gaze determination
• remote physiological monitoring for high stress
• human speech recognition, natural language understanding to support human-robot interaction and high level control by operators and peers
• language translators, human-machine dialogue
• use of avatars and human face representations to facilitate expressive communication.
Ideally this research will be integrated with a computational cognitive model to enable the robot to
• make inferences regarding the roles, potential threats, goals and communications of humans
• provide the operator with high level communication with the robot and dismounted persons.
Propose the design and prototype development of a system capable of peer-to-peer embedded human-robot interaction that exploits a critical combination of sensing, communication and reasoning, as described in Phase I.
Development and testing of a fully mobile, self-contained robot capable of interaction with both cooperative and uncooperative humans in a peer-to-peer relationship which provides tactically significant stand-off capabilities to human operators, while retaining high level control.
Commercialization potential includes:
• operation of mobile robots in hazardous conditions (Chemical / Biological / Nuclear) in populated areas
• police tactical operations such as hostage situations
• urban search and rescue where communication with subjects is important
• use of robot sentries in high threat conditions medical triage under hazardous conditions.
1. C. Breazeal (2003) “Towards sociable robots”, T. Fong, (ed) Robotics and Autonomous Systems, 42(3-4), pp. 167-175.
2. J. L. Burke, R.R. Murphy, M.D. Coovert, and D.L. Riddle (2004), “Moonlight in Miami: A field study of human-robot interaction in the context of an urban search and rescue disaster response training exercise.” Human-Computer
Interaction, vol. 19, pp. 85-116.
3. Trafton, J.G., Schultz, A.C., Cassimatis, N.L., Hiatt, L.M., Perzenowski, D., Brock, D.P., et al (2006) Communicating and collaborating with robotic agents. In R. Sun (Ed.), Cognition and Multi-Agent Interaction: From Cognitive Modeling to Social Simulation (pp. 252-278). New York, NY: Cambridge University Press.
4. Trafton, J.G., Cassimatis, N.L., Bugajska, M.D., Brock, D.P., Mintz, F.E. and Shultz, A.C. (2005) “Enabling effective human-robot interaction using perspective-taking in robots.” IEEE Transactions on Systems, Man and
Cybernetics, 35(4), 460-470.
5. Oviatt, S.L., Cohen, P.R., and Wang, M.Q. (1994) “Toward interface design for human language technology: Modality and structure as determinants of linguistic complexity”, Speech Communication 15, 3-4, 1994, pp. 283-300.
6. Perzanowski, D, Shultz, A., Adams, W., Marsh, E. and Bugajska, M. (Jan./Feb. 2001) “Building a Multimodal Human-Robot Interface”, IEEE Intelligent Systems, Vol. 16, no. 1, IEEE Computer Society, pp. 16-21.
Tanaka Business School Imperial College Exhibition Road London SW7
25th - 29th September 2006Launch Evening:25th September, 6pm - 9pm
This exhibition will feature work from six artists who use computers in diverse ways to generate their artworks. Their techniques include the following:
Generative artThe modeling of evolutionary processes generates complex, organic and beautiful shapes and images.
Computers as an artistic mediumComputational techniques open up exciting new opportunities for multi-media art generation, and the combination of computer generated art with traditional mediums.
Simulation of the artistic processSimulating artistic and creative processes enables computers to generate human-like artworks.
William Latham is Professor of Creative Technology at Leeds Metropolitan University, and research fellow at Goldsmiths College. With the mutator program, he was an early innovator in the field of evolutionary art, and his organic artworks and films were shown worldwide. After making a similar impact in the games industry, he has recently returned to the world of computer generated art.
Penousal Machado is a researcher at the Creative Systems group, at the Centre for Informatics and Systems, University of Coimbra, Portugal. His research focuses on the development of artificial artists and computer aided creativity. He has developed the NeVar program, an evolutionary art tool which can also produce artworks autonomously. He is a organiser of the EvoMusArt conference series.
Simon Colton is a lecturer in Computing at Imperial College, and creative director of Machine Creations Ltd. He investigates the notion of computational creativity in domains such as mathematics, bioinformatics and painting. His TripTych program simulates painting, and using this, he has set up the Craft By Numbers service, where customers can paint stylistic versions of their digital photos.
“At Craft by Numbers, we supply high quality paint by numbers kits with which you paint a masterpiece. The best part about this is that the painting is based on a photograph of your choice. The results are amazing: great looking paintings of people and places which really mean something to you.”
If one of the following sounds familiar, then a Craft by Numbers painting kit is perfect for you:
• You love doing paint by numbers, but you want something a little different.
• You want to capture forever in an artistic way a particular person or occasion special to you.
• You have a great photo which you think would make a wonderful painting.
• You are looking for a fun and creative pastime for yourself or a loved one.
• You are looking for a personalized and unique gift for a friend or family member.
• You want to get into the wonderful world of painting
Craft by Numbers is different to traditional paint by numbers that you buy in shops. Here are a few things which make our service unique:
• You no longer have to paint make-believe scenes. The painting you create is a faithful, artistic version of the photograph you send to us.
• Our painting kits contain high quality art products from leading art supplier Daler-Rowney. No more little pots of paint, Craft by numbers is much more like normal painting.
• There are no unsightly numbers to paint over: we provide a cheat-sheet which describes exactly which colours go where.
• You mix acrylic paints in a palette, to ensure that you get just the right colours to fit your photograph. Each painting requires the mixing of around 30 colours, and we supply simple instructions for doing this.