Home | News Updates | In This Issue | Company Directory | About Us | Subscribe | Contact Us
  BATTLESPACE update & BATTLESPACE email alert service is sponsored by Northrop Grumman
   Free e-mail exhibition news service, Register here!
Click read  more about BATTLESPACE TECHNOLOGIES

NEW TECHNOLOGIES  

ISSN 1416-300X Volume 12, Issue 2, July 2010

Page | 1 | 2 | 3 |

Web Page sponsor Oxley Developments
www.oxleygroup.com

Oxley Group Ltd
Oxley specialises in the design and manufacture of advanced electronic and electro-optic components and systems for air, land and sea applications within the military sector. Established in 1942, Oxley has manufacturing facilities in the UK and USA and enjoys representation worldwide. The company’s products include night vision and LED lighting, data capture systems and electronic components. Oxley has pioneered the development of night vision compatible lighting. It offers a total package incorporating optical filters, equipment modification, cockpit and external lighting along with fleet wide upgrade services including engineering, installation, support, maintenance and training. The company’s long experience of manufacturing night vision lighting and LED indicators, coupled with advances in LED technology, has enabled it to develop LED solutions to replace incandescent and fluorescent lighting in existing applications as well as becoming the lighting option of choice in new applications such as portable military hospitals, UAV control stations and communication shelters.


THE RISE OF THE MACHINES
By Eur Ing Paul Parkinson, Senior Systems Architect, Aerospace & Defence, Wind River

Abstract

This article considers the reasons for the increased adoption of military robots in recent years, the evolution and exploitation of underlying technologies, and potential future applications and the issues involved.

Military robots used to belong to the realms of science fiction popularised by movies such as the Terminator series, but science fiction is now becoming reality. Since the end of the Cold War era, there has been a significant change in conventional engagement requirements, with the long-established scenarios of infantry, armoured divisions and air forces being deployed against a similarly equipped enemy replaced by asymmetric warfare scenarios involving coalition forces against irregular forces in remote terrain and urban warfare environments.
Armed forces have realised that traditional forces need to adapt in order to be successful in these new theatres of operation. In particular, they are presented with the significant challenge of being able to successfully engage with a dispersed irregular enemy, which may use guerrilla tactics, in urban settings on remote terrain. This represents a new type of threat which can present a higher risk to dismounted forces, and the loss of life which may be incurred may also have political consequences.

For these scenarios, an alternative which reduces the threat to soldiers is needed. Robots have been used extensively for bomb-disposal over the years, and they have the potential for adaptation for mine-clearance and urban house clearance roles. In addition, they could be used in an advance reconnaissance role. Being the first exposed to potential threats, they reduce the risk to soldiers. Unmanned air vehicles (UAVs) are another form of robot which can be used in a reconnaissance role, and they have been widely used to provide broader situational awareness of the battlefield area.

A second class of robot is needed, a surveillance robot, to perform a similar function for soldiers, perhaps even with the potential to outperform their human counterparts, and without ever tiring. An obvious example of this is nighttime patrols and sentry duty, where human vision is impaired compared to daylight and the night vision receptors can play tricks in the dark, leading to false alarms, or worse, nondetection of threats. A good example of the use of this technology in the civil domain is QinetiQ’s Tarsier robot[1], which checks airport runways for foreign object debris, to avoid crashes such as that of the Concorde in 2000.
The third class of robot, which is somewhat controversial, is the armed combat robot, for use in offensive combat operations. This is viewed as being important for high-risk and high-threat engagement scenarios and could be deployed instead of troops to reduce the risk of casualties. It could also be used where compromise or capture could be politically sensitive. In the airborne environment, the Predator-B UCAV extends the capabilities of the Predator UAV through the addition of ground strike capability and can be deployed on operations where there is an unacceptable level of risk to a military aircraft and its pilot.

The Technologies

In recent years, there has been a significant increase in the development of military robots, due to the increasing need referred to previously but also due to the technical maturity and availability of technologies which enable robots to perform their basic functions. These technologies include computer vision, communications and autonomy.

Computer Vision

The potential of robots is evident by the state-of-the-art computer vision technologies which provide them with capabilities which far exceed their human counterparts. The use of multispectral electro-optic and infrared (EO/IR) sensors by high-altitude long-endurance (HALE) UAVs for reconnaissance and target identification is already widely known. These technologies can also be used in ground-based robots, along with microwave-length sensors, to provide enhanced vision even in low-visibility conditions, including darkness and fog. This enables robot vision to achieve greater sensitivity, range and endurance than in humans.

Communications

The advances in recent years have made real-time communication between a remotely deployed robotic system and a distant command-and-control centre feasible. In particular, there has been a migration from stovepipe systems to open architecture systems based on the Internet protocol (IP), in conjunction with the security framework provided by IPsec and more recently IPv6, which can be used in conjunction with high-grade encryption and the appropriate protocols to enable the following:
* Confidentiality: Ensuring that the message remains secure during transit across the network
* Authentication: Ensuring that the sender of the message is who they claim to be
* Integrity: Ensuring that the message remains intact during transit

This ensures that the command data link between the control ground station and the UAV remains secure and that hostile forces cannot take control of the UAV through a “man in the middle attack” and divert it from its mission. In addition, as the bandwidth of IP-based networks has also increased dramatically in recent years, this has made streaming of high-resolution imagery in real-time feasible, which avoids the issues associated with local storage and enables decisions to be made by human operators in real-time.

Autonomy

The challenge for robotic and unmanned systems is processing the captured imagery data and making intelligent decisions either semi-autonomously or fully autonomously. In the case of remotely controlled robots and UAVs, the processing requirement is not too great because the decision-making based on the processed data is undertaken by the human operator, provided that there is a suitable communications link available. In cases where this link is not available, a greater degree of autonomy is required.

The robot or UAV can perform some decision-making by itself (e.g., where to go/fly) based on a preplanned profile (as in the case of the Predator, Watchkeeper [2] and other UAVs). It achieves this by making use of geographic and terrain information databases, GPS and inertial navigation systems, enabling it to follow preprogrammed routes to obtain new imagery, then return to base where it can either land autonomously or under remote control when in range of a ground control station. However, in the case of unexpected scenarios, it results in a greater challenge for the robot. For example, the UAV needs to determine whether ground vehicles belong to coalition or enemy forces, because it may be difficult to perform image matching against a target identification database in all cases.

The Future

Robotic systems have exploited recent advances in technology, but we are now reaching the point where the evolution of autonomous systems could have greater consequences in the near future. This is due to the advent of two classes of robot: remotely controlled semi-autonomous armed robots, and fully autonomous armed robots. The distinction here is due to whether the robot has the ability to decide if and when to fire, or whether the decision needs to be made by a human counterpart. These issues of safety and control will need to be considered carefully. This is likely to be a contentious subject in coming years because armed robots contradict Asimov’s Laws of Robotics [3], and despite the portrayal of these laws in movies such as I, Robot [4], these laws cannot be implemented in software in a simple manner.

In addition, the reasoning ability of robotic systems does yet not match human capability. For example, the armed military robots which have been developed to act as sentries on the Korean border have the ability to autonomously decide whether to open fire on approaching humans [5], but they do not appear to have the capability to determine whether an approaching soldier is a friend or a foe, or similarly, if approached by a child with a toy gun which does not pose a threat. So this system currently only makes sense when used in the demilitarized zone (DMZ).

In order for armed autonomous systems to be deployed safely, the reasoning capabilities of robots will need to increase dramatically to the point where they possess processing power of commander data, whose computational power is 60 trillion operations per second [6] which arguably exceeds that of humans. Some academic experts are claiming that this is many years away. In the meantime, the development and deployment of armed robots should be undertaken by strict adherence to safety-critical software development standards to ensure that the probability of software error leading to an unintended fatality is minimized.

The widespread deployment of robots in significant roles currently undertaken by humans will depend on the social acceptability of robots in these roles. A recent survey by the Royal Academy of Engineering [7] indicated that this might not present an insurmountable barrier because 28 percent of respondents said that they would prefer to see androids flying aircraft. Although many of the automated systems on modern civil aircraft help to reduce the pilot’s workload, and autopilot systems are capable of landing an aircraft without human intervention under certain conditions, they are not sufficiently intelligent to deal with every eventuality in the way that a human pilot can, and it is not clear whether the respondents to the survey were aware of this, which could, of course, affect their views.
In any case, it is expected that autonomous aircraft would be used for cargo transport initially, and some estimates have suggested that they could be in service by 2050. But if the geometric increase in computing performance continues indefinitely, could this be achieved sooner than expected?


References
[1] Patrick Beasley, “Tarsier,” IET Seminar on the Future of Civil Radar, (15 June 2006). http://www2.theiet.org/Events/Beasley_Full_Presentation.pdf
[2] “Watchkeeper Tactical UAV System,” Army Technology. http://www.army-technology.com/projects/watchkeeper/
[3] “Three Laws of Robotics,” Wikipedia. http://en.wikipedia.org/wiki/Three_Laws_of_Robotics
[4] I, Robot, (2004), Internet Movie Database. http://www.imdb.com/title/tt0343818/
[5] “A Robotic Sentry for Korea's Demilitarized Zone,” IEEE Spectrum. http://www.spectrum.ieee.org/robotics/military-robots/a-robotic-sentry-for-koreas-demilitarized-zone
[6] “Commander Data,” Memory Alpha. http://memory-alpha.org/en/wiki/Data
[7] “Robot Gets Green Light to Manage England Team,” Royal Academy of Engineering. (5 February 2008). http://www.raeng.org.uk/news/releases/shownews.htm?NewsID=439 This survey revealed that 48 percent said that they would be happy to see a robot manage the England football team, but it is unclear whether this would improve the team’s chances in the 2012 World Cup.

About the Author
Eur Ing Paul Parkinson is a senior systems architect with Wind River, working with customers in the aerospace and defence sectors in the UK and across EMEA. His professional interests include information security (InfoSec), Integrated Modular Avionics (IMA) and Intelligence Surveillance Target Acquisition Reconnaissance (ISTAR) systems. He blogs on A&D industry issues on the Wind River website at http://blogs.windriver.com/parkinson.

Page | 1 | 2 | 3 | Next page

Advertisers
Visit the Oxley website
Visit the Paradigm website
Hobson Industries
Go to MILLBROOK websitte
COMTECH
TRaC


Design by Net Resources.