System Bits: Feb. 28

Computer bots; nano sandwich; agricultural robots.

popularity

Software robots have fights lasting years
According to University of Oxford and Alan Turing Institute researchers, editing bots on Wikipedia undo vandalism, enforce bans, check spelling, create links and import content automatically, whereas other non-editing bots mine data, identify data or identify copyright infringements — sometimes with unpredictable consequences.

The team looked at how much disruption the bots caused on Wikipedia sites, seeing how they interacted on 13 different language editions from 2001 to 2010, and found that bots interacted with one another, whether or not this was by design, which led to unpredictable consequences. Their research paper concluded, interestingly, that bots are more like humans than might be expected as they appear to behave differently in culturally distinct online environments. The paper says the findings are a warning to those using artificial intelligence for building autonomous vehicles, cyber security systems or for managing social media. It suggests that scientists may have to devote more attention to bots’ diverse ‘social life’ and their different cultures.

Although the online world has become an ecosystem of bots, the knowledge of how they interact with each other is still rather poor, the researchers assert. And although bots are automatons that do not have the capacity for emotions, bot to bot interactions are unpredictable and act in distinctive ways. They found that German editions of Wikipedia had fewest conflicts between bots, with each undoing another’s edits 24 times, on average, over ten years. This shows relative efficiency when compared with bots on the Portuguese Wikipedia edition, which undid another bot’s edits 185 times, on average, over ten years. Bots on English Wikipedia undid another bot’s work 105 times, on average, over ten years, three times the rate of human reverts.

The findings show that even simple autonomous algorithms can produce complex interactions that result in unintended consequences – ‘sterile fights’ that may continue for years, or reach deadlock in some cases.

Importantly, the researchers also found that the same technology leads to different outcomes depending on the cultural environment. For instance, an automated vehicle will drive differently on a German autobahn to how it will through the Tuscan hills of Italy. Similarly, local online infrastructure that bots inhabit will have some bearing on how they behave and their performance

The bottom line: more study is needed.

2D optoelectronic hybrids simulated
In order to expand the optoelectronic properties of magnesium oxide, Rice University researchers have modeled a nanoscale sandwich, by putting two slices of atom-thick graphene around nanoclusters of the super conductive material.

Nanoclusters of magnesium oxide sandwiched between layers of graphene make a compound with unique electronic and optical properties, according to researchers at Rice University who made computer simulations of the material. (Source: Rice University)

Nanoclusters of magnesium oxide sandwiched between layers of graphene make a compound with unique electronic and optical properties, according to researchers at Rice University who made computer simulations of the material. (Source: Rice University)

Rice materials scientist Rouzbeh Shahsavari and his colleagues built computer simulations of the compound and found it would offer features suitable for sensitive molecular sensing, catalysis and bio-imaging. They believe this work could help researchers design a range of customizable hybrids of two- and three-dimensional structures with encapsulated molecules.

Agricultural robots
In order to identify the genetic traits in plants likely to produce the greatest yields, a semiautonomous robot being developed by University of Illinois, Cornell University and Signetron researchers may soon roam agricultural fields to gather and transmit real-time data about the growth and development of crops.


A robot under development at the University of Illinois automates the labor-intensive process of crop phenotyping, enabling scientists to scan crops and match genetic data with the highest-yielding plants. Agricultural and biological engineering professor Girish Chowdhary, right, is working on the $3.1 million project, along with postdoctoral researcher Erkan Kayacan.
(Source: University of Illinois)


A robot under development at the University of Illinois automates the labor-intensive process of crop phenotyping, enabling scientists to scan crops and match genetic data with the highest-yielding plants. Agricultural and biological engineering professor Girish Chowdhary, right, is working on the $3.1 million project, along with postdoctoral researcher Erkan Kayacan.
(Source: University of Illinois)

The researchers said the agricultural robot was inspired by the autonomous rovers used to search collapsed buildings, and other dangerous environments, is propelled on continuous tracks, or miniature tank treads that allow it to navigate through dry or muddy fields. It is guided using GPS, and a laptop computer.


Traveling between the crop rows, the robot uses hyperspectral, high-definition and thermal cameras, weather monitors and pulsed laser scanners to capture phenotypic information – such as the stem diameter, height and leaf area of each plant – and assess environmental conditions, such as the temperature and moisture content of the soil.

The team pointed out that the underlying technologies – the algorithms, the mechanical design and the human-robot interaction devices that provide robustness – are useful in many other industries, including defense, surveillance and scientific exploration. A prototype is expected within a few years, and the robot is expected to be on the market by 2021.