As 21th century enters the realm of futurism, society is slowly seeing artificial intelligence being assimilated into our modern technology. Huge companies such as BOEING and Tesla are incorporating autonomous driving into commercial airlines and automobiles; both these companies want to take another huge leap forward by taking humans out of the equation and integrating artificial intelligence capable of making the same decisions as drivers or pilots.
Artificial Intelligence takes the data analytics ranging from descriptive to diagnostic to predictive and lastly prescriptive. The descriptive analytics is part of the hindsight stage where researchers receive information about the event and occurred. Once receiving news on the event, researchers then use the data gathered such as from data discovery, data-mining, and correlations to understand why and how did this event happen; the diagnostic analytics happen in the insight stage. The last stage is called the “Foresight” stage. It includes the predictive and prescriptive analytics which projects outcomes that will occur and how to utilize A.I in making these outcomes happen on a consistently positive basis.
The article of “age-related kinematics” discusses the differences of age that is influenced by task difficulty, target size, and movement amplitude. The data given in these articles will be analyzed and explained on how it can be further developed with Artificial Intelligence to enhance efficiency and safety. After I will explain how HEX uses data analytics to develop HEX’s Dynamic Evacuation System and improve on safety and security measure by incorporating A.I.
In the research article called “Age-Related Kinematic Difference as Influenced by Task Difficult, Target Size, and Movement Amplitude” conducted C. Ketcham, R. Seidler, A. Gemmert, and G. Stelmach compares fifteen older adults with a median age of 68 and 15 young adult with a median age of 23 to assess three specific hypotheses.
1. Task Difficult Comparison: If movement slowing can be attributed to a global-information-processing deficit, then total movement time will increase differentially in older adults compared with young adults as index of difficult increases. Older adults will product differentially lower peak velocities and longer deceleration phases with more inflections across all movements regardless of parameters.
2. Influence of Target Size: If movement slowing is specifically related to a deficit in accuracy control, we hypothesized that when amplitude is held constant, older adults will show a much greater increase in movement durations relative to young adults when target size is decreased. Older adults will produce movements wit longer deceleration phases containing more inflections earlier in the velocity profile compared with your controls.
3. Influence of movement amplitude: If movement slowing conversely is specifically related speed control, we hypothesized that when target size is held constant, older adults will demonstrate a greater increase in movement duration relative to young adults when amplitude is increased. Older adults will produce movements with lower velocities compared with young adults.
The results shown that older adults respond to change in task difficulty by making adjustments in response to accuracy and amplitude meaning that older adults are unable to effectively propel their limb to the target in a single step causing their movements to be less and smooth slower. In addition, when accuracy was not a factor in an aiming movement, older adults produced movement velocities that were similar to young adults, so ultimately adults has the ability to produce forces to propel them forward but the ability to control the forces is determined by the accuracy requirement. Older adults have difficulty decelerating their movement and resulting in slower movement to have better coordination.
With the data analysis taken from this study, engineers and researchers can utilize data and implications to later teach the computer to read the behaviors of humans. For example, if a smart learning device was employed into a residential home, it can analyze the movement speed and precisions of a human to get a better understanding of its kinetic movement; if and when the movement does slow down or completely stops, the A.I. can use historical data collected from the resident to see if it is a complete anomaly. Other features must also be incorporated such as infrared cameras that can measure temperature of the patient, auditory frequency meters to notice in change in breathing and/or speaking habits, and video analyzation of distortion in movement. All these features combined with the data analysis measuring the kinematic movement of residents or patients can define the 1) age 2) physical condition 3) mental condition and 4) consciousness of them. A.I. and big data analytics can play a vital role in determining the health conditions of humans and will allow people to respond more appropriately and efficiently to these types of situations.
HEX’S Dynamic Evacuation
HEX uses the research conducted by the University of Greenwich called “Active Dynamic Signage System: A Full-Scale Evacuation Trial” and built upon the data to create HEX’s Dynamic Evacuation System. In the article, it states that efficient evacuations within terminals or public transportations can be constrained by the complex nature of the buildings. Even though emergency signage systems are widely used in many facilities and they are used to facilitate evacuees, only 38% of people ‘see’ the conventional emergency signage in an emergency situation. When switching from the conventional emergency signage to an active emergency signage (a light that demonstrates a flashing lights directing people which way to walk) the percentage doubles to a 76%.
HEX's Dynamic Fire Evacuation System is a software based system that analyzes the hazard potential of the fire area and calculates the most optimal and safest egress route out of the building. HEX’s Dynamic Emergency Signage lights interpret the calculations within the evacuation panel and then visually directs evacuees which way to exit towards. HEX’s Dynamic Evacuation is a smart system and intuitively calculates the fire area and arranges the egression route itself; if and when the fire does spread to a different location and our sensors will then be notified and our evacuation panel will re-route to a different path if necessary.
Although HEX’s Dynamic Evacuation System uses the A* Algorithm (link to explain the use of A* Algorithm) as a tool to assist evacuees out the premise, other features can also be added to the evacuation system to learn behavior mechanism of humans to allow a safer passageway out. One of HEX’s projects is to combine a voice evacuation system with our dynamic evacuation system. This new system will not only allow us to visually and verbally communicate with evacuees but also raise awareness of the correct pathway to use. An example of integrating A.I intelligence into our system is having the machine understand the temperature and humidity of the room if there is actually a fire occurring. After extensively examining to room – to lower false-positives-- the artificial intelligence can automatically shut all ventilation and doors around the fire area to prevent the fire from growing and spreading quicker. Software engineers can take fire response techniques and teach machines to help respond to fire situations quicker without any hesitation.
We are seeing huge companies like BOEING and Tesla incorporating Artificial Intelligence into their commercial airlines and automobiles. Soon society will be taking the huge leap forward by incorporating A.I. into our daily jobs. Artificial Intelligence cannot be done without any data analytics because data discovery and mining is the backbone in creating a well-structured and formulated system. These analytics can be drawn from research papers and countless hours of testing to prove its variability; the data can be used to teach artificial intelligence how to understand the correct procedure and intuitively judge in the situation with the right response.
Galea, E.R., H. Xie, D. Cooney, and L. Filippidis. "ACTIVE DYNAMIC SIGNAGE SYSTEM: A FULL- SCALE EVACUATION TRIAL." London, The University of Greenwich, 2015.
George Anadiotis - http://www.zdnet.com/article/data-to- analytics-to-ai-from-descriptive-to-predictive-analytics/
Ketcham, Caroline J., Rachael D. Seidler, Arenda W. Van Gemmert, and George E. Stelmach. "Age-Related Kinematic Differences as Influenced by Task Difficulty, Target Size, and Movement Amplitude." Vol. 57, 2001, pp. 54-64.