«ARL-SR-0327 ● JUNE 2015 US Army Research Laboratory Visualizing the Tactical Ground Battlefield in the Year 2050: Workshop Report by Alexander ...»
The workshop viewed such collaborative robot/human behaviors as highly likely to be common in 2050. As one participant pointed out, the well-known video of robot quadrotors performing James Bond theme (https://www.youtube.com/ watch?v=_sUeGC-8dyk) already gives a vivid suggestion of what a collaborative attack by an intelligent team of munitions might look like. The feasibility of this capability is further supported by what is believed to be the scalability of swarm behaviors and by the relative ease of manufacture and deployment of swarmsuitable (smaller) robots. Because collective, collaborative behaviors are critical for survival in a contested environment saturated with sensors and robo-munitions, sufficient research and development (R&D) efforts will have been required to achieve these kinds of swarming or teaming capabilities.
There are a broad range of means that could be used to defeat a swarm or team of robots. Physical nets of other physical obstacles could be used to deny or shield some areas from robots. Robots could be equipped to wire cutters or other means to penetrate a net or a shield. Alternatively, some of the robot team members could be assigned suicide missions where they would explode themselves in order to breech a net or shield. Swarms or teams of robots could be kinetically attacked using area weapons or a high volume of fire. A counter to this would be to disperse the swarm and concentrate it only for a short period of time when a combined effect is needed.
Another means of countering a collaborative robot formation would involve attacking their ability to communicate with one another and/or sense their environment. This could involve jamming, EM pulse (EMP), a cyber-hack, or an attack on robot communications capability. Robot networks could be made more agile and less vulnerable to such attacks by being equipped with several alternative modes of communications, making it difficult to jam or otherwise attack of all these modes simultaneously.
A different approach to countering swarms or teams of robots involves reverseengineering a robot team’s behavior to create a predictive model that would be helpful in designing a plan to prevent the robot team from achieving its mission.
The ability of an adversary to understand and predict robot behaviors could be countered by introducing a stochastic element into the algorithms that determine robot behaviors.
3.2.3 Dynamic Hacking and Spoofing Hacking involves the ability to penetrate a system and change it in some way (change code, scramble data, insert malware). The results of a hack attack could vary from a one-time compromise to creating a vulnerability that can be exploited at will. Spoofing can be considered to be “behavior hacking.” The intent of behavior hacking is to influence behavior by altering the information upon which behaviors are based. Both hacking and spoofing can be used in “agile” attacks that dynamically, as a function of the circumstances and conditions, select the effects that are created.
Dynamic hacking and spoofing is likely to be a prominent feature of the tactical environment circa 2050 because 1) the “attack surface” of robot teams and swarms is large, which makes these forms of attack highly attractive; and 2) increased interest and attention is being paid to research that would enables such attacks. This includes efforts to automate reverse engineering and intelligent vulnerability analysis.
The designers of robots are well aware of the likelihood of hacking and spoofing attacks, and hence, will endeavor to design individual robots, robot networks, and robot-to-robot and robot-to-human communications capabilities with this in mind.
Efforts will therefore be made to increase their internal security, harden their communications, and make robot sensing and processing less vulnerable to hacking and spoofing. The addition of hardware-based security (hard points, kernels of trust) will serve to make them less vulnerable to hacking (but may make them unable to adapt as attack vectors change). In addition to these countermeasures, decentralization (employing mission command) will reduce the reliance on communications and large numbers of heterogeneous robots will make them more difficult to attack or attrite. Other countermeasures would involve increasing the agility of individual robots by enabling dynamic repurposing and/or building in an override feature that could be exercised by human controllers. As a last resort, responses to the degrading or disabling of robot formations should be included in contingency plans.
3.2.4 Super Humans The principal Army unit operating in 2050 will be mixed human-robot teams. To enable humans to partner effectively with robots, human team members will be enhanced in a variety of ways. These super humans will feature exoskeletons, possess a variety of implants, and have seamless access to sensing and cognitive enhancements. They may also be the result of genetic engineering. The net result is that they will have enhanced physical capabilities, senses, and cognitive powers.
The presence of super humans on the battlefield in the 2050 timeframe is highly likely because the various components needed to enable this development already exist and are undergoing rapid evolution.
There are a variety of means available to countering the presence of super humans, ranging from simply overwhelming them with numbers of less expensive robots, area munitions, and directed EMP. Given that the powers super humans possess will in large part depend upon communications (as least internal to the superhuman) and computer processing, hacking and spoofing attacks are also an option.
To counter these attacks on high value super humans, such humans could be protected/shielded by robot clouds or force fields. Given their capabilities, super humans would employ tactics that exploit their unique qualities and capabilities.
They would be designed and outfitted to be able to continue to function even when damaged and, as a last resort, “battle plan” contingencies designed to operate without effective super humans would be standard operating procedure.
3.2.5 Directed Energy Weapons The promise of DEWs was described in a 2007 Defense Science Board report of the same name. This report states that “directed energy continues to offer promise as a transformational ‘game changer’ as the DOD encounters new asymmetric and disruptive threats, while facing increasingly sophisticated traditional challenges.” Several DEW technologies that have shown promise have also presented significant challenges. These include high power micro and millimeter wave, and lasers of various kinds (solid-state, chemical, fiber), both airborne and ground. However, in the past decade, these technologies have received increased attention and, as a result of the progress that has subsequently been made on both the technologies themselves and the sources needed to power these weapons, workshop participants consider it likely that a variety of these weapons will be employed in 2050.
There are a number of ways that the intended targets of DEWs can counter their attacks. Targets can be designed with a variety of characteristics and techniques to reflect, refract, and disperse the energy directed at them. These include surface features and contours and the use of active defenses like chaff or dust. The use of multi-spectral decoys can be effective, as well as can the ability to rapidly maneuver out of the effective envelope of the attack. Intended targets can also use cover and concealment to avoid attacks. Attacking the DEW power source can also be an effective counter strategy.
3.2.6 Force Fields Force fields consist of particles, energy, or waves that destroy, cripple, or otherwise interfere with objects that attempt to penetrate them. Given the variety, precision, and lethality of the weapons and the ubiquitous nature of the sensors that will be found on the battlefield of 2050, considerable attention will be devoted to developing force fields that can both help protect easy to locate assets and track high value targets. Workshop participants believe that sufficient progress will be made to make it likely that force fields will be employed in 2050 for some of the same reasons given for DEWs. In addition, workshop participants felt that force fields would be developed because they are seen as a counter to DEWs.
Furthermore, the decreasing utility and cost-effectiveness of armor make force fields attractive alternatives.
A counter to the deployment of a force field is to use faster, bigger rounds and/or shoot from closer in, because the shorter the range, the easier it is to defeat the force field. Along these lines increasing the volume of fire, employing barrages or fire swarms could overwhelm a force field. Depending upon the nature of the force field, adjustments to DEWs or other weapons may help them better penetrate.
Alternatively attacking the power source and/or hacking the systems that are involved could also be effective counters.
3.2.7 Reliable Power Sources Power is needed to operate all of the robots, the technologies that make humans “super,” DEWs, as well as force fields. Some of these 2050 battlefield capabilities require a great deal of energy to function effectively. The reliable supply of this energy is essential for mission effectiveness.
Workshop participants felt that as a result of the considerable attention that is being paid to the development of improved power sources and power storage (lighter, more efficient, more cost-effective, faster recharge) that power would not be the limiting factor in the ability to deploy the technologies discussed in this report.
Confidence in this assertion was increased as workshop participants also envisioned more sources of power on the battlefield than are found today, including the
an ability to tap into the power infrastructure indigenous to the battlefield • an ability to hijack an adversary’s sources of power • A variety of attacks on power sources is likely on the battlefield of 2050. These include direct kinetic attacks, EMP, cyber, DEW, and force fields to prevent wireless power transmission, and employment of power “leeches.” Counters to these attacks would include treating power sources as high value targets and defend them accordingly (harden, conceal). Deploying ample backup power sources is also an attractive option.
3.2.8 Other Observations While most of the discussion involved 1 of the 7 developments discussed above,
the participants made a number of additional observations:
invisibility cloaks •
high-fidelity, predictive modeling of influence operations • collecting and modeling the data about individual adversaries •
4. Summary and Conclusion The diverse set of workshop participants painted a vivid picture of the battlefield of 2050, one that brought reality more in line with the science fiction and fantasy the public is accustomed to viewing in the cinema and reading about. A time traveler from today would be immediately taken with the “over-crowding” of the battlefield of 2050 populated by all manner of robots, robots that greatly outnumber human fighters, and robot-looking humans. Not immediately apparent to the time traveler, but critical in determining which of the adversaries would possess the decisive edge, would be the capabilities and autonomy possessed by the armies of virtual robots, the “intelligent” programs and processes to 1) collect, process, and disseminate information to develop situational awareness; 2) direct and manage collections of robots that were engaged in executing C2, combat-support functions, as well as combat missions; and 3) undertake a full range of defensive and offensive cyber operations.
A critical challenge of the mid-21st century will involve successfully managing and integrating the collections, teams, and swarms of robots that would act independently or collaboratively as they undertook a variety of missions including the management and protection of communications and information networks and the provision of decision-quality information to humans. Success in this aspect of command and control would depend upon developing new C2 concepts and approaches, in particular, developing and fielding an effective hybrid cognitive architecture that leverages the strengths of artificial intelligence and human intelligence to go along with the development of new robotic, communications, information, and systems technologies. From the various observations of workshop participants, the traditional balance between offense and defense may shift as it becomes more difficult for the defense to keep up.
INTENTIONALLY LEFT BLANK.
Appendix A. Workshop Participant Presentations List of Symbols, Abbreviations, and Acronyms