Abnormal response of chaotic system based on artificial intelligence

Through artificial intelligence systems and autonomous agent devices, sensor data in chaotic environments are analyzed, pseudo-Brownian motion and non-Brownian motion are distinguished, and abnormal situations are responded in a timely manner, which solves the problem of difficult to detect and deal with potential hazards in chaotic environments in the existing technology, and improves the problem of improving the problem of difficult to detect and handle potential hazards in chaotic environments. The stability and security of the system are achieved.

CN113632105BActive Publication Date: 2024-11-19MORGAN STANLEY SERVICES GROUP INC
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202080025222.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Priority Date
2019-01-31
Filing Date
2020-01-30
Publication Date
2024-11-19
Estimated Expiration
2040-01-30

AI Technical Summary

Technical Problem

The existing technology is difficult to effectively detect and respond to abnormal situations in chaotic environments, especially in complex systems, such as network congestion, traffic flow and weather changes, and it is difficult to distinguish between pseudo-Brownian motion and non-Brownian motion, resulting in the failure of potential harm. Deal with timely manner.

Method used

An artificial intelligence system and method is designed to receive sensor data from a chaotic environment through a central server and autonomous proxy device, analyze pseudo-Brown changes in the data, determine the expected range of variables, and trigger autonomous response measures when the variables are out of range. to mitigate potential hazards.

Benefits of technology

It realizes timely detection and response to abnormal situations in chaotic environments, reduces the impact of potential hazards, and improves the stability and security of the system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN113632105B_ABST
    Figure CN113632105B_ABST
Patent Text Reader

Abstract

A system for detecting and responding to anomalies in a chaotic environment, the system comprising one or more autonomous agent devices and a central server. The central server receives a first set of sensor readings from a remote electronic sensor during a first time window, the sensor readings recording pseudo-Brownian changes in one or more variables in the chaotic environment; determines an expected range of the variable during a second time window after the first time window based on the first set of sensor readings; receives a second set of sensor readings from the remote electronic sensor during the second time window recording the change in the variable; determines that a variable is not within an expected range based on the second set of sensor readings; and causes the autonomous agent device to attempt to mitigate a potential hazard indicated by a variable being outside the expected range.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to artificial intelligence methods and systems, and more particularly to methods and systems for artificial intelligence to analyze sensor data from a chaotic environment and to influence the environment. Background Art

[0002] In 1827, Robert Brown discovered some of the first evidence for the molecular theory of matter by observing how pollen grains in water were pushed back and forth in unpredictable ways by collisions bouncing off invisible water molecules. As the quality of observational devices improved, it was discovered that although all fluids consist of individual molecules that travel in straight lines when unimpeded by collisions with other molecules, all fluids are highly chaotic due to the countless collisions that constantly change the momentum of these molecules.

[0003] Brownian motion - the property of the movement of individual particles in this chaotic fluid system - has been extensively analyzed in the mathematical fields of statistical physics, applied physics, and topology. In addition, many non-fluid systems have been found or hypothesized to be modelable as if they undergo Brownian motion, such as the movement of stars in a galaxy, or the changes in the value of investment assets in the market. Summary of the invention

[0004] The present invention discloses an artificial intelligence system for detecting and responding to anomalies in a chaotic environment, the artificial intelligence system including one or more autonomous agent devices and a central server, the central server including a processor and a non-transitory memory. The memory stores instructions that, when executed by the processor, cause the processor to perform the following: receiving a first set of sensor readings from one or more remote electronic sensors during a first time window, wherein the sensor readings record pseudo-Brownian changes in one or more variables in the chaotic environment; determining an expected range of one or more variables during a second time window after the first time window based on the first set of sensor readings; receiving a second set of sensor readings from one or more remote electronic sensors during the second time window that records changes in one or more variables; determining that one of the one or more variables is not within an expected range (and therefore, whether it is assumed that it is unlikely to be the result of pseudo-Brownian motion, but rather non-Brownian motion) based on the second set of sensor readings; and causing one or more autonomous agent devices to attempt to mitigate potential hazards indicated by one variable being outside the expected range.

[0005] Also disclosed is an artificial intelligence method for detecting and responding to anomalies in a chaotic environment, the artificial intelligence method comprising: receiving a first set of sensor readings from one or more remote electronic sensors during a first time window, wherein the sensor readings record pseudo-Brownian changes in one or more variables in the chaotic environment; determining an expected range of one or more variables during a second time window after the first time window based on the first set of sensor readings; receiving a second set of sensor readings from one or more remote electronic sensors during the second time window that record changes in one or more variables; determining, based on the second set of sensor readings, that one of the one or more variables is outside an expected range; and causing one or more autonomous agent devices to attempt to mitigate potential hazards indicated by a variable being outside the expected range. BRIEF DESCRIPTION OF THE DRAWINGS

[0006] Figure 1 depicts a computing system for receiving distributed sensor readings of a chaotic environment and directing an agent in response to changes in the chaotic environment;

[0007] Figure 2 Possible projections of expected behavior in chaotic systems with pseudo-Brownian motion are depicted;

[0008] Figure 3 Describes the methods used by AI systems to process incoming sensor data and guide agents in cluttered environments;

[0009] Figure 4 depicts an actual graph showing changes in sensor readings and expected range over a period of time when the expected range is repeatedly recalculated after a new time window ends;

[0010] Figure 5 depicts an actual graph showing sensor readings over a period of time as a variable experiences a sudden decrease and returns to normal; and

[0011] Figure 6 A general purpose computing device for performing the various features described above is depicted. DETAILED DESCRIPTION

[0012] Figure 1 Depicted is a computing system for receiving distributed sensor readings of a chaotic environment and directing an agent in response to changes in the chaotic environment.

[0013] A chaotic system with multiple discrete "active" entities that influence each other or "passive" entities within the system can be described as exhibiting pseudo-Brownian motion of those active or passive entities. The simpler the behavior of the entities being described, the more likely it is that the overall behavior of the entities in the environment will match the overall behavior of a true Brownian motion environment; the behavior of air or water flow can be better approximated than the behavior of birds in a flock, and the behavior of birds can be better approximated than the behavior of people in a crowd. The motion also does not have to be physical matter, rather than values; for example, the number of data connections maintained by a network device can vary in response to the independent behavior of users (each user chooses to initiate a connection) and the dependent behavior of users (each user chooses to terminate a connection because network congestion renders the network useless). In this example, impatient and patient users entering and leaving the network at different times can be abstracted as if high-speed and low-speed molecules were pushing each other in a fluid, causing the network congestion level itself to undergo pseudo-Brownian motion.

[0014] Now go to Figure 1 The central server 100 receives sensor data from multiple remote electronic sensors 105 in the chaotic environment via a network 110 and transmits instructions to multiple electronic computing device agents 115 via the network 110.

[0015] The network 110 may be, for example, the general Internet, a local wireless network, an Ethernet network or other wired network, a satellite communication system, or any other means of connecting the sensors 105 to the central server 100 and the central server to the agent 115 to enable data transmission. Furthermore, the network 110 may not be a single network as shown, but rather multiple separate networks; for example, the central server 100 may have multiple proximal sensors 105 to which it is attached via wired connections, multiple nearby sensors 105 to which it is connected via a Wi-Fi network, and multiple extremely remote sensors 105 to which it is connected via satellite. The connections may avoid the use of a network altogether and use direct wired or wireless transmissions to send data to and from the central server 100. As shown in FIG. Figure 1 As shown, arrows show the intended direction of data flow to and from the network.

[0016] Sensor 105 may be any type of electronic sensor that registers data from a chaotic environment external to central server 100. Example sensor types for particular implementations may include, but are not limited to, cameras, thermometers, GPS tracking devices, or other geo-location devices, sensors of motion / distance / acceleration / orientation, or communications modules that receive electronic data communications from a source.

[0017] The agent 115 may be any form of computing device capable of causing changes in the chaotic environment or in another "real world" system affected by the chaotic environment. For example, the agent 115 may be a computing device that is used to trigger a physical alarm, pilot a drone aircraft or autonomous vehicle, route network traffic, generate a message for display on a physical device associated with a human user, or perform other actions associated with a "smart home appliance" or other automated system.

[0018] Several possible pairings of sensors 105 and agents 115 to achieve particular purposes are described below.

[0019] In one example implementation, sensors 105 may include thermometers at various locations inside a computer or server room, and agents 115 may be climate control fans or air conditioners. The system may prevent physical damage to computers and improve computer performance by using sensor data to determine whether an observed rise in temperature may be a random change or a substantial problem, such as a blocked fan grill or an overheating component. In response to a problem, additional cooling systems may be activated or increased in power, automatic alarms may be triggered, and human operators of the system may be notified of unexpected temperature changes.

[0020] In another example implementation, sensor 105 may be an outdoor thermometer that measures the temperature of an entire city or landscape. The system may prevent inaccurate weather report data by using data from the thermometer to determine that changes in observed temperature are not weather related, but are human-caused (such as the operation of an exhaust fan for heating air near the thermometer, or the presence of a campfire near the thermometer), and automatically notify the weather service that data from a particular sensor should be excluded or treated with caution.

[0021] In another example embodiment, sensors 105 may include GPS trackers on multiple animals, or cameras that record the location of animals in a nature reserve. Abnormal behavior in the movement of animals may indicate the presence of predators, environmental hazards, or unwanted predators in the nature reserve that cause animals to move in a manner that is different from the usual pattern of grazing and movement within the herd. Agents 115 may be aerial drones sent to observe the situation and report data back to park management, or personal devices or alarms used by park management to alert them that they should send someone to investigate the situation.

[0022] In another example implementation, the sensors 105 may include GPS trackers in multiple cars. Once a "drift" variable is included to normalize the calculations by subtracting the average speed of the traffic flow during all calculations, the traffic flow can be modeled as a pseudo-Brownian motion of vehicles "bouncing" off each other by braking when they get too close to each other. The agents 115 may include automated systems for managing traffic (such as traffic lights or additional traffic lanes that can be opened or closed), the autonomous vehicles themselves, or personal devices or alarms used by drivers, police, or other first responders. In response to traffic jams, road hazards, unexpected traffic volumes, or other issues that affect the natural flow of traffic, various automated systems can be directed to change behavior in a way that reduces traffic impedance, and human actors can also be alerted to the problem to inform their options.

[0023] In another example embodiment, the sensors 105 may include one or more cameras that track the location of multiple people in a public space, such as a shopping mall. The agents 115 may include a fire alarm or other alarm system, a cleaning robot, or a computer system capable of generating messages to direct human workers. The behavior of a person who deviates significantly from "walking around" or heading in a general direction while avoiding contact with others may indicate a passive hazard such as a liquid spill or a noxious odor, which influences people to take a longer path to avoid the hazard, or an active hazard such as a fire or a person with a weapon, from which people are fleeing. In response, the system may automatically attempt to determine the cause of the hazard, trigger an alarm if necessary, and direct automated or human resources to resolve the hazard.

[0024] In another example implementation, the sensor 105 may include a firewall or router at the edge of a computer network, reporting the number of incoming network packets, and the agent 115 may include a server, router, or firewall in a server cluster. The system can detect the beginning of a denial of service (DDOS) attack by distinguishing an increase in network traffic from natural variations in incoming traffic, and the system can activate more servers to handle the attack or cut off the inflow of network traffic until the attack stops.

[0025] Similarly, the sensor 105 may report the total use of resources on the device, such as the load on the CPU (central processing unit), network card, or memory. The agent 115 may include a kernel or operating system process for shutting down or throttling software that is using excessive or increasing proportions of system resources by determining that an increase in resource usage indicates a program error or malware design rather than a natural change during the expected use of the software. The operating system process can then respond by automatically terminating the software, throttling the resources available to the software, sandboxing the software to isolate it from other system components, or generating a warning for a human user of the software.

[0026] In another example implementation, the sensor 105 may include a tracker of network traffic flows to a particular resource, such as a page for a particular movie on a streaming service or a page for a particular product on an e-commerce site, and the agent 115 may include a server within the streaming service or an automated warehousing element associated with the e-commerce website. The system may detect the beginning of a surge in popularity for a given resource or item (such as triggered by a celebrity endorsement or other unexpected cultural or economic change) and begin directing a content delivery network to distribute the electronic resource more widely, or may direct an automated warehousing element to move the storage location of an item for more efficient shipment from a warehouse.

[0027] In another example embodiment, the sensor 105 may include a device that reports the current buy or sell price of one or more assets at a stock exchange or other market. The agent 115 may include a computing device that can send a buy or sell order to the market or recall a buy or sell order from the market, or may include a firewall device that can prevent such a computing device from successfully sending a buy or sell order to the market. In response to a market anomaly, the trader's ability to trade may be automatically stopped by the computing device itself, or the trader may be notified of the anomaly so the trader can proceed more cautiously.

[0028] The determination that the observed change in the sensor value indicates a problem that must be addressed by one of the agents 115 is based at least in part on the observation of the sensor value outside of an expected range (e.g., Figure 2 The expected range is determined by artificial intelligence analysis of past sensor values ​​and current sensor values ​​(as shown in Figure 3 as shown and described further below).

[0029] Figure 2 Possible projections of the expected behavior of sensed values ​​in a chaotic system with pseudo-Brownian motion are depicted.

[0030] Figure 2 205 and predicted future sensor values ​​210. Expected range 215, including upper range 216 and lower range 217, constrains predicted future sensor values ​​210, as the additional time available for random or pseudo-random changes in sensor values ​​expands over time (axis 220) as uncertainty increases.

[0031] The expected range 215 represents a confidence interval such that if the change in sensor value is a pseudo-Brownian motion and remains driven by pseudo-Brownian motion without active interference, then future observed sensor values ​​are less likely than a certain threshold to exceed the interval. For example, in one embodiment, the sensor value may be expected to not leave the expected range 215 with a probability of 95%, absent some unaccounted-for factor affecting the underlying chaotic environment sensed by the sensor 105. In another embodiment, the expected range 215 may be wider and represent a probability of 99% or 99.9% or even higher that the sensor value will not leave the range in the absence of active interference. Examples of these varying confidence intervals are shown below. Figure 4 In the description.

[0032] like Figure 2 As shown, the observed and predicted sensor values ​​205 and 210 are single scalar values, so the cross-section perpendicular to each point along the time axis 220 of the expected range 215 is a vertical line, and the expected range 215 is essentially a two-dimensional wedge with the current point. In other embodiments, the sensor values ​​205 and sensor values ​​210 can be multi-dimensional. For example, if two sensor values ​​are measured in tandem, the expected range 215 will be a three-dimensional pyramid or cone, where its current point has a two-dimensional cross-section relative to the time axis 220 and represents the expected range of possible pairs of two sensor values. Higher-dimensional expected ranges for values ​​of multiple sensors can be extrapolated based on the same principles (although not easily shown).

[0033] Although boundaries 216 and 217 of expected range 215 are depicted here as straight line boundaries, curved boundaries, or other nonlinear boundaries (or non-planar boundaries / non-hyperplane boundaries in higher dimensional implementations), expected range 215 may be limited based on the characteristics of the chaotic environment or the sensitivity of the system to anomalies.

[0034] Figure 3 A method for an artificial intelligence system to process input sensor data, generate Figure 2 and guiding the agent in a chaotic environment in response to subsequent sensor readings leaving an expected range.

[0035] At the beginning of each time window in a series of time windows, one or more sensor readings are received from the sensor 105 by the central server 100 (step 300). The time windows can be adjusted for a particular implementation and represent any time period from a few minutes when measuring the movement of an animal to a few seconds when measuring the temperature of a system or the movement of a person to a fraction of a minute when monitoring vehicle movement, network traffic, resource usage, or asset price fluctuations.

[0036] Any detected drift terms (i.e., all sensor readings experiencing a systematic shift in one direction, as may occur in some embodiments) are removed from the data (step 305). For example, as described above, the positions of vehicles in a traffic jam may not only change in response to each other and hazards, but will also be constantly changing at approximately the average speed of the traffic flow. Therefore, the sensor readings, including the total speed of the vehicles, should be normalized by subtracting the drift term, the average traffic speed. After normalization, some vehicles will have negative speeds relative to the flow, while other vehicles will have positive speeds, as one would expect from observing positive and negative speeds of fluid particles in a container in a given direction. Similarly, in an upward trending market following good economic news, it may be necessary to isolate and remove the drift term in order to determine whether a particular asset price change is abnormal or in line with the market trend.

[0037] Other actions may be taken to normalize data from different types of sensor readings, or as otherwise desired in a given implementation.

[0038] Then, looking at the normalized sensor readings from the previous time window, the double power change is calculated (step 310) according to the following equation:

[0039]

[0040] where s(t) is the sensor reading at time window t and lowercase sigma represents the volatility estimator, which is the simulated standard deviation in the sensor data calculated by the double power variation.

[0041] Based on the calculated standard deviation and the desired risk tolerance for the particular application, an expected range is calculated for one or more upcoming time windows assuming that the variable is undergoing pseudo-Brownian motion (step 315). For example, a 90% confidence interval that the variable is undergoing pseudo-Brownian motion is typically calculated by determining a range that is approximately plus or minus four sigma from the average sensor value, while a 99.9% confidence interval may correspond approximately to plus or minus six sigma from the average sensor value.

[0042] More specifically, a confidence interval that the change in a variable is pseudo-Brownian rather than non-Brownian can be constructed by choosing L. The ratio of sensor readings that should be considered as possibly abnormal to sigma is such that P(abnormal) = exp(-exp((C(n)-L) / S(n))), where C(n) is equal to ((2log n) 1.5 / c)-((log pi+log log n) / 2c(21og n) 0.5), n is the total number of windows and hence the index of the current window, c is the constant sqrt(2 / pi), and S(n) is equal to l / c(2log n) 0.5 function.

[0043] If each incoming sensor reading is within the confidence interval (step 320), no action is taken except to return to receiving new sensor readings in the next time window and recalculating the expected range (repeating steps 300 to 315). If an anomalous sensor reading is detected, the central server 100 can generate a message for transmission to one or more of the autonomous agents 115 (step 325), which can then be further configured to minimize harm based on the application of the anomaly detection system (step 330). Regardless of the actions of the autonomous agents 115, the central server 100 continues to review the sensor data to determine if additional anomalies exist, or whether further sensor data appears to indicate a "new normal" for variability in the chaotic system and subsequent adjustments to the expected range, such as Figure 4 shown.

[0044] Figure 4 Depicted is an actual graph showing the change in sensor readings and expected range over a period of time as the expected range is repeatedly recalculated after the end of a new time window.

[0045] The expected maximum value 216 and minimum value 217 represent the 99.9% confidence interval if the probability is affected only by pseudo-Brownian motion and not by external non-Brownian motion. Figure 4 The sensor readings plotted with points at 400, 401, 402, and elsewhere in will fall within this confidence interval.

[0046] The expected maximum value of 216 and minimum value of 217 are not as Figure 2 The linearity shown is based on the recalculation of past and current data over time, so that Figure 4 Effectively representing a small slice of each single projection occurring immediately after the current time in that projection.As the change in sensor readings 400 increases or decreases during the time window, the range between expected maximum and expected minimum values ​​at those time points widens and narrows accordingly.

[0047] At two points in time (near sensor reading 401 and sensor reading 402), the volatility in sensor reading 400 increases significantly, such that sensor reading 401 and sensor reading 402 are above confidence interval 216 and confidence interval 217, respectively, and three additional readings at approximately the same time are below confidence interval 216 and confidence interval 217, respectively. These anomalous data points strongly indicate some kind of active interference in the normal function of the chaotic environment, and may be the basis for the actions of autonomous agent 115.

[0048] Because the outliers may but are unlikely to be sensor readings that are not affected by the system anomaly, and because other later sensor readings may still be affected by the disturbed chaotic system, the confidence intervals are significantly expanded after outliers 401 and 402 to prevent false alarms from subsequent sensor readings that may represent normal chaotic behavior following such a disturbance in the system.

[0049] Figure 5 Depicted is an actual graph showing sensor readings over a period of time as a variable experiences a sudden decrease and returns to normal.

[0050] During most of the time period when the three variables 500 were experiencing changes, the L ratio remained below 4 at almost all times because the variables 500, while noisy, did not change significantly within any small time window. However, at time 501, there was a sudden and sharp drop in one of the three variables, causing the L ratio to spike to approximately 10, indicating a very high probability that the change was not random noise in the change of variables or another example of pseudo-Brownian motion. Therefore, a system monitoring changes in these variables should trigger and take automatic actions as needed.

[0051] At time 502, another sudden change occurs, causing the L ratio to increase even further to a value of approximately 14. However, in this case, despite the incredibly high L value, the system should not have to take action because the L value has been skewed by the lower and more stable cycle of measured variable 500. The jump at time 502 may represent a return to the mean and may represent the end of an anomalous influence on variable 500, rather than a new secondary influence further distorting it.

[0052] The system's decision to take or inhibit automatic action may be influenced not only by statistical analysis of incoming sensor data over short time windows, but also by sensor data over long windows or supplemental sensor data or information feeds that do not directly report variable 500 values, but rather report information that may affect variable 500 values ​​and may help analyze whether a second disturbance in the variable value is a reversion to the mean. Selecting a longer time window for data analysis can help avoid overreaction of the system to sudden changes and recovery / self-correction, but there is also a risk of making the system less responsive to changes that will not self-correct. Empirical testing of a given implementation can help determine the window size and number of consecutive windows for optimizing the tradeoff between unnecessary action and undesirable inaction.

[0053] Figure 6 1 is a high-level block diagram of a representative computing device that can be used to implement the various features and processes described herein, for example, the functionality of the central server 100, the sensor 105, or the autonomous agent 115. The computing device can be described in the general context of instructions executable by a computer system, such as program modules executed by the computer system. Generally, program modules can include routines, programs, objects, components, logic, data structures, etc. that perform particular tasks or implement particular abstract data types.

[0054] like Figure 6 As shown, the computing device is shown in the form of a special purpose computer system. The components of the computing device may include (but are not limited to) one or more processors or processing units 900, a system memory 910, and a bus 915 that couples various system components including the memory 910 to the processor 900.

[0055] Bus 915 represents one or more of any of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, and a processor or local bus using any of a variety of bus architectures. By way of example and not limitation, these architectures include an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MCA) bus, an Enhanced ISA (EISA) bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus.

[0056] Processing unit 900 can execute computer programs stored in memory 910. Any suitable programming language can be used to implement the routines of a particular embodiment, including C, C++, Java, assembly language, etc. Different programming techniques such as procedural or object-oriented can be used. The routines can be executed on a single computing device or multiple computing devices. In addition, multiple processors 900 can be used.

[0057] Computing devices typically include a variety of computer system readable media. Such media can be any available media accessed by the computing device, and such media include volatile and non-volatile media, removable and non-removable media.

[0058] The system memory 910 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 920 and / or cache memory 930. The computing device may also include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 940 may be provided for reading from and writing to a non-removable non-volatile magnetic medium (not shown and generally referred to as a "hard drive"). Although not shown, a disk drive for reading and writing from a removable non-volatile disk (e.g., a "floppy disk") and an optical drive for reading or writing from a removable non-volatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In this case, each may be connected to the bus 915 via one or more data media interfaces. As will be further described and described below, the memory 910 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the various embodiments described in the present disclosure.

[0059] By way of example and not limitation, a program / utility 950 having a set (at least one) of program modules 955, as well as an operating system, one or more application software, other program modules, and program data may be stored in memory 910. Each of the operating system, one or more application software, other program modules, and program data, or some combination thereof, may include an implementation of a networking environment.

[0060] The computing device may also communicate with one or more external devices 970, such as a keyboard, pointing device, display, etc.; one or more devices that enable a user to interact with the computing device; and / or any device that enables the computing device to communicate with one or more other computing devices (e.g., a network card, a modem, etc.). Such communication may occur via an input / output (I / O) interface 960.

[0061] In addition, as described above, the computing device can communicate with one or more networks such as a local area network (LAN), a general wide area network (WAN), and / or a public network (e.g., the Internet) via network adapter 980. As depicted, network adapter 980 communicates with other components of the computing device via bus 915. It should be understood that, although not shown, other hardware and / or software components may be used in conjunction with the computing device. Examples include (but are not limited to) microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data archival storage systems, etc.

[0062] The description of various embodiments of the present invention has been presented for the purpose of illustration, but is not intended to be exhaustive or limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terms used herein are selected to best explain the principles of the embodiments, practical applications, or technical improvements to technologies found in the marketplace, or to enable those of ordinary skill in the art to understand the various embodiments disclosed herein.

[0063] The present invention may be a system, method and / or computer program product at any possible level of technical detail integration. The computer program product may include a computer-readable storage medium (or multiple media) having computer-readable program instructions for causing a processor to perform various aspects of the present invention.

[0064] Computer readable storage medium can be a tangible device that can retain and store instructions used by an instruction execution device. Computer readable storage medium can be, for example, but not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. A non-exhaustive list of more specific examples of computer readable storage medium includes the following: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device such as a punch card or a convex structure in a groove on which instructions are recorded, and any suitable combination of the foregoing. Computer readable storage medium as used herein should not be interpreted as a temporary signal itself such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagated by a waveguide or other transmission medium (e.g., a light pulse by a fiber optic cable), or an electrical signal transmitted by a wire.

[0065] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to a corresponding computing / processing device, or downloaded to an external computer or external storage device via a network such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical transmission fibers, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing device / processing device receives the computer-readable program instructions from the network, and forwards the computer-readable program instructions to be stored in a computer-readable storage medium in the corresponding computing device / processing device.

[0066] The computer-readable program instructions for performing the operation of the present invention can be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, configuration data of integrated circuits, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, and process programming languages ​​such as "C" programming language or similar programming languages. Computer-readable program instructions can be executed completely on the user's computer, partially on the user's computer, executed as an independent software package, partially on the user's computer and partially on a remote computer, or completely on a remote computer or server. In the latter case, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, using an Internet service provider through the Internet). In some embodiments, the electronic circuit including, for example, a programmable logic circuit, a field programmable gate array (FPGA) or a programmable logic array (PLA) can execute computer-readable program instructions to personalize the electronic circuit by utilizing the state information of the computer-readable program instructions, and then perform various aspects of the present invention.

[0067] Various aspects of the present invention are described herein with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer programs according to various embodiments of the present invention. It should be understood that each frame of the flowchart illustration and / or block diagram and the combination of frames in the flowchart illustration and / or block diagram can be implemented by computer-readable program instructions.

[0068] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device create a device for implementing the functions / actions specified in one or more boxes of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium, which can instruct the computer, programmable data processing device, and / or other device to function in a specific manner, so that the computer-readable storage medium having the instructions stored therein includes an article of manufacture, which includes instructions for implementing various aspects of the functions / actions specified in one or more boxes of the flowchart and / or block diagram.

[0069] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing device, or other apparatus, so that a series of operational steps are performed on the computer, other programmable device, or other apparatus to produce a computer-implemented process, so that the instructions executed on the computer, other programmable device, or other apparatus implement the functions / actions specified in one or more boxes of the flowchart and / or block diagram.

[0070] Flowcharts and block diagrams in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention.In this respect, each frame in the flow chart or block diagram can represent a module, section or part of an instruction, which includes one or more executable instructions for realizing the specified logical function.In some alternative embodiments, the function pointed out in the frame may not occur in the order pointed out in the figure.For example, two frames shown in succession can actually be performed substantially simultaneously, or these frames can sometimes be performed in reverse order, depending on the function involved.It will also be noted that the combination of the frames in each frame of the block diagram and / or flow chart illustration and the block diagram and / or flow chart illustration can be realized by a system based on special-purpose hardware, which performs a specified function or action or performs a combination of special-purpose hardware and computer instructions.

Claims

1. An artificial intelligence system for detecting and responding to anomalies in a chaotic environment, comprising: one or more autonomous agent devices; as well as A central server comprising a processor and a non-transitory memory storing instructions that, when executed by the processor, cause the processor to: receiving a first set of sensor readings from one or more remote electronic sensors during a first time window, wherein the sensor readings record pseudo-Brownian motion changes of one or more variables in the chaotic environment; determining, based on the first set of sensor readings, an expected range of the one or more variables during a second time window following the first time window based at least in part on a bi-power variation calculated from the sensor readings in the time portion within the first time window; receiving a second set of sensor readings from the one or more remote electronic sensors recording changes in the one or more variables during the second time window; determining, based on the second set of sensor readings, that one of the one or more variables is outside the expected range; and causing the one or more autonomous agent devices to attempt to mitigate a potential hazard indicated by the one variable being outside the expected range; The determination of the expected range is based at least in part on a predetermined probability, such that in the absence of active interference in the chaotic environment, the sensor reading will fall within the expected range at least with the predetermined probability.

2. The system according to claim 1, wherein: The expected range is determined based at least in part on a double power variation calculated from normalized sensor readings in a time portion within the first time window, and the expected range is calculated for one or more upcoming time windows.

3. The system according to claim 1, wherein: The one or more autonomous agent devices attempt to mitigate the potential harm by preventing network messages from being transmitted over the network.

4. The system according to claim 1, wherein: The one or more autonomous agent devices attempt to mitigate the potential hazard by activating physical devices that act on the chaotic environment.

5. The system according to claim 1, wherein: The one or more autonomous agent devices attempt to mitigate the potential hazard by driving the autonomous vehicle to a specific location.

6. The system according to claim 1, wherein: The one or more autonomous agent devices attempt to mitigate the potential harm by generating messages for receipt by a human user.

7. The system according to claim 1, wherein: The one or more autonomous agent devices attempt to mitigate the potential hazard by activating an alarm visible or audible to a human user.

8. The system according to claim 1, wherein: The instructions, when executed by the processor, further cause the processor to expand the expected range based at least in part on the one or more variables that are not within the expected range.

9. The system according to claim 1, wherein: The instructions, when executed by the processor, further cause the processor to reduce the expected range based at least in part on a third set of sensor readings indicating that the one or more variables are within the expected range.

10. An artificial intelligence method for detecting and responding to anomalies in a chaotic environment, comprising: receiving a first set of sensor readings from one or more remote electronic sensors during a first time window, wherein the sensor readings record pseudo-Brownian motion changes of one or more variables in the chaotic environment; determining, based on the first set of sensor readings, an expected range of the one or more variables during a second time window following the first time window based at least in part on a bi-power variation calculated from the sensor readings in the time portion within the first time window; receiving a second set of sensor readings from the one or more remote electronic sensors recording changes in the one or more variables during the second time window; determining, based on the second set of sensor readings, that one of the one or more variables is outside the expected range; and One or more autonomous agent devices are caused to attempt to mitigate a potential hazard indicated by the one variable being outside of the expected range.

11. The method according to claim 10, wherein: The expected range is determined based at least in part on a double power variation calculated from normalized sensor readings in a time portion within the first time window, and the expected range is calculated for one or more future time windows.

12. The method according to claim 10, wherein: The determination of the expected range is based at least in part on a predetermined probability, such that in the absence of active disturbances in the chaotic environment, sensor readings will fall within the expected range with at least the predetermined probability.

13. The method according to claim 10, wherein: The one or more autonomous agent devices attempt to mitigate the potential harm by preventing network messages from being transmitted over the network.

14. The method according to claim 10, wherein: The one or more autonomous agent devices attempt to mitigate the potential hazard by activating physical devices that act on the chaotic environment.

15. The method according to claim 10, wherein: The one or more autonomous agent devices attempt to mitigate the potential hazard by driving the autonomous vehicle to a specific location.

16. The method according to claim 10, wherein: The one or more autonomous agent devices attempt to mitigate the potential harm by generating messages for receipt by a human user.

17. The method according to claim 10, wherein: The one or more autonomous agent devices attempt to mitigate the potential hazard by activating an alarm visible or audible to a human user.

18. The method of claim 10, further comprising expanding the expected range based at least in part on the one or more variables not within the expected range.

19. The method of claim 10, further comprising narrowing the expected range based at least in part on a third set of sensor readings indicating that the one or more variables are within the expected range.