Prognostics driven decision making

JP2022084023A5Active Publication Date: 2024-12-03PALO ALTO RESEARCH CENTER INC
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
JP2021191178
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Priority Date
2020-11-25
Filing Date
2021-11-25
Publication Date
2024-12-03
Estimated Expiration
2041-11-25

AI Technical Summary

Technical Problem

Existing systems lack effective methods to predict and respond to the remaining useful life (RUL) of assets, leading to inadequate proactive maintenance and potential system failures during missions, which can disrupt operations and increase costs.

Method used

A system and method for monitoring asset health, estimating RUL, and automatically initiating corrective actions such as failsafe modes, controller adaptations, system reconfigurations, or maintenance based on RUL categories to minimize disruption and optimize operational readiness.

Benefits of technology

Enables proactive maintenance planning and reflexive system protection, reducing operational disruptions and costs by tailoring responses to the predicted RUL, ensuring timely maintenance and system resilience.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 00000000_0000_ABST
    Figure 00000000_0000_ABST
Patent Text Reader

Abstract

To perform decision making for appropriate corrective measures by estimating a remaining useful life.SOLUTION: A method monitors a health of at least one asset. A remaining useful life (RUL) of the asset is estimated based on the monitoring. The RUL is classified into categories including those shorter than a time to complete a current mission and those longer than the time to complete the current mission. One or more corrective measures are automatically executed during the current mission if the RUL is classified as being those shorter than the time to complete the current mission. The corrective measures include either one or more of: initiating a fail-safe mode; adapting a controller of the one or more assets; reconfiguring a system; and adjusting the current mission of the one or more assets. Maintenance is scheduled for after the current mission of at least one asset if the RUL is classified as being those longer than or equal to the time to complete the current mission.SELECTED DRAWING: Figure 1
Need to check novelty before this filing date? Find Prior Art

Description

Background Art

[0001] Predictive estimation can be useful for estimating the remaining useful life of various types of equipment. By estimating the remaining useful life, equipment operators can make decisions based on information regarding appropriate corrective actions.

Summary of the Invention

Problems to be Solved by the Invention

[0002] Some embodiments include a method that includes monitoring the health of at least one asset. The remaining useful life (RUL) of the at least one asset is estimated based on the monitoring. The RUL of the asset is classified into categories including those shorter than the time to complete the current mission and those longer than the time to complete the current mission. One or more corrective actions are automatically executed during the current mission if the RUL is classified as being less than the time to complete the current mission. The corrective actions include one or more of initiating a failsafe mode, adapting a controller of one or more assets, reconfiguring the system, and adjusting the current mission of one or more assets. If the RUL is classified as being greater than or equal to the time to complete the current mission, maintenance is scheduled after the current mission of the at least one asset.

[0003] Some embodiments include a system comprising a processor and memory that stores computer program instructions that, when executed by the processor, cause the processor to perform actions. The operation includes monitoring the health of at least one asset. The remaining useful life (RUL) of at least one asset is estimated based on the monitoring. The RUL of an asset is categorized into two groups: shorter than the time to complete the current mission and longer than the time to complete the current mission. One or more corrective actions are automatically performed during the current mission if the RUL is classified as shorter than the time to complete the current mission. The corrective actions include one or more of initiating failsafe mode, adapting the controllers of one or more assets, reconfiguring the system, and adjusting the current mission of one or more assets. If the RUL is classified as longer than the time to complete the current mission, maintenance is scheduled after the current mission of at least one asset.

[0004] Some embodiments include a non-temporary computer-readable medium storing computer program instructions, which, when executed by a processor, cause the processor to perform an action. The action includes monitoring the health of at least one asset. The remaining useful life (RUL) of at least one asset is estimated based on the monitoring. The RUL of an asset is categorized into two groups: shorter than the time to complete the current mission and longer than the time to complete the current mission. One or more corrective actions are automatically performed during the current mission if the RUL is classified as shorter than the time to complete the current mission. The corrective actions include one or more of the following: initiating failsafe mode, adapting the controllers of one or more assets, reconfiguring the system, and adjusting the current mission of one or more assets. If the RUL is classified as longer than the time to complete the current mission, maintenance is scheduled after the current mission of at least one asset.

[0005] The above summary is not intended to describe each embodiment or all implementations. A more complete understanding will be revealed and understood by referring to the following detailed description and claims in conjunction with the attached drawings. [Brief explanation of the drawing]

[0006] [Figure 1] Figure 1 shows a method for determining one or more corrective actions to be taken based on a Rule of Unlawful Use (RUL) according to embodiments described herein. [Figure 2] Figure 2 shows a more detailed flowchart for determining one or more corrective actions to be taken based on the RUL according to the embodiments described herein. [Figure 3] Figure 3 shows exemplary results of a simulation of a hypothetical asset having 10 subsystems / components that fail at various times with different RUL values, according to the embodiments described herein. [Figure 4-1] Figure 4-1 shows a more detailed diagram of each component according to the embodiments described herein, where each component is represented by its RUL and operating state. [Figure 4-2] Figure 4-2 shows a more detailed diagram of each component according to the embodiments described herein, where each component is represented by its RUL and operating state. [Figure 4-3] Figure 4-3 shows a more detailed diagram of each component according to the embodiments described herein, where each component is represented by its RUL and operating state. [Figure 5] Figure 5 shows a block diagram of a system in which embodiments of the specifications described herein can be carried out.

[0007] Drawings are not necessarily to scale. Similar numbers used in drawings refer to similar components. However, it should be understood that the use of numbers to refer to components in a given drawing is not intended to limit components in another drawing labeled with the same number. [Modes for carrying out the invention]

[0008] The approaches described herein relate to the field of making predictions based on the remaining useful life (RUL) of several types of components, such as pumps, valves, transformers, engines, medical systems, structural members, and batteries. All equipment deteriorates and will eventually fail if corrective actions are not taken. Such predictions are particularly important in the field of prescriptive maintenance (RxM), where this predictive information is used to provide decisions that minimize operational disruption, thereby reducing costs compared to unscheduled, reactive maintenance. For example, if the predicted RUL (i.e., time to failure) is only a few milliseconds away, scheduling maintenance may not be an appropriate response. Instead, the system may be configured to reflexively protect itself and then address the recovery function later. However, if the prediction is, for example, that an asset will fail within three weeks, there may be sufficient time to perform maintenance actions that will help correct the problem. Embodiments described herein involve determining one or more actions based on the predicted RUL.

[0009] A system may have many subsystems and components used in the mission to perform their respective functions. These degrade based on usage and may also experience failures, which manifest as sudden changes in the healthy trajectory of a subsystem or component. These failures are stochastic in nature and can occur at any given time. The prior probability of a failure occurring is typically known, but the time at which they occur is not. The remaining life decreases more rapidly when failures are present compared to when they are not. Different actions should be taken depending on the nature of the failure and the time to failure (which prediction estimates as remaining useful life). If the failure reduces the RUL to an imminent point of failure, the system may reflexively protect itself and enter fail-safe mode. This may sacrifice operational readiness but protect the larger mission goals. The system may be configured to return assets to a perfect state where appropriate corrective actions are possible. Such corrective actions may include maintenance.

[0010] If the RUL (Reason for Failure) is greater than the impending failure but still does not allow for maintenance, system reconfiguration may be appropriate depending on the scope of prediction. System reconfiguration may involve using directly or indirectly redundant subsystems that can at least partially restore operational capability. Reconfiguration can be achieved for various types of systems, including mechanical, electrical, or other types. The extent to which this type of reconfiguration can be achieved depends on the modularity and architecture of the system. For complex systems, finding partial and indirect redundancy may be achieved by examining numerous possible configurations and comparing the expected functionality.

[0011] Another viable reconfiguration may be controller adaptation. Controller adaptation may involve changing gains or other controller settings. This should aim to alleviate some stress on the components and, ideally, result in some recovery of remaining life. Finding the appropriate settings can be a complex task depending on the complexity of the control. One way to address this task is to explore different control settings from a bank of controllers. Another way would be to perform an optimization of the control settings so that the desired functionality encompasses not only fulfilling the mission objectives but also recovering remaining life. A larger RUL may allow for modifications to the mission plan to address the lack of availability of a particular subsystem (which may be temporary). Replanning or rescheduling a task is a search or optimization problem and typically requires more time depending on the overall mission complexity. Replanning a mission results in reduced use of a failed component (if it can be rendered completely unusable). Where possible, the failed component should be repaired.

[0012] Finally, if the RUL is even greater, maintenance can be scheduled taking into account various operational and logistical constraints. Typically, many competing objectives need to be considered in order to best address this task. These include minimizing the impact on operational objectives, repairing at the lowest cost, minimizing downtime, making the best use of repair equipment, and any potential impact on secondary objectives (such as factory workload planning).

[0013] Maintenance measures are performed based on the recommendations of the maintenance optimization tool, and, if no recommendation is made at that time, based on the urgency of the estimated remaining life. That is, components may experience a reduction in remaining life even if the failure did not cause a sudden decline. Components may be scheduled to be repaired to prevent further failure.

[0014] Figure 1 shows a method for determining one or more corrective actions to be taken based on RUL, according to embodiments described herein. The health of at least one asset is monitored, for example, via one or more sensors (110). Depending on the configuration, the health of the assets can be monitored continuously. In some cases, the health of the assets may be monitored at predetermined discrete intervals. According to various embodiments, the health of assets can be monitored based on a pre-predicted RUL. According to various implementations, it can be determined based on monitoring whether a failure has occurred in one or more assets.

[0015] The RUL of at least one asset may be estimated based on monitoring (120). For example, the RUL may be based on an estimate of future damage to a component. Estimates may employ fundamentally different approaches to provide an estimate of future damage to a component. For example, one or more estimators may operate from first principles of the physics of the system and / or fault propagation with respect to given operating and environmental conditions to provide an estimate of future damage. Alternatively or additionally, one or more estimators may operate from a model based on empirical data obtained through numerous experiments, or from a collection of data trajectories of past failures, to provide an estimate of future damage. Predicting the RUL of an asset is described in detail in Owner Common U.S. Patent Application No. 16 / 717,649, which is incorporated herein by reference in its entirety.

[0016] The RUL for at least one asset may be classified based on the length of the RUL (130). For example, RULs may be classified as time less than the time to complete the current mission and time less than the time to complete the current mission. It should be understood that the term “mission” can refer to and / or be considered a mission that includes subtasks. Generally, this mission is a goal-driven activity that can be as diverse as the flight of an aircraft or the processing of materials in a manufacturing plant.

[0017] One or more corrective actions may be performed automatically if the RUL is classified as having less time to complete the current mission (140). If the RUL is classified as having more time to complete the current mission, maintenance may be scheduled after the current mission of at least one asset (150). According to embodiments described herein, maintenance may include any measures required to maintain or improve the health of the asset. For example, maintenance may include one or more of the following: repair, replacement, refurbishment, cleaning, lubrication, etc.

[0018] According to various embodiments, one or more corrective actions include initiating failsafe mode, adapting the controllers of one or more assets (e.g., adjusting the settings of one or more controllers), reconfiguring the system, and adjusting the current missions of one or more assets. According to various configurations, adjusting the current missions includes replanning and / or rescheduling one or more mission tasks. Replanning one or more mission tasks may include, for example, changing the order in which mission tasks are completed and / or eliminating one or more unnecessary mission tasks.

[0019] RUL may be classified into any number of categories. For example, RUL may be classified into the first, second, third, and fourth categories. RUL may be classified into the first category if the remaining RUL is less than or equal to the first threshold; it may be classified into the second category if the remaining RUL is greater than the first threshold but less than or equal to the second threshold; if the remaining RUL is greater than the second threshold but less than or equal to the third threshold, at least one asset's RUL may be classified into the third category; and if the remaining RUL is greater than the third threshold, at least one asset's RUL may be classified into the fourth category. One or more of the first, second, and third thresholds may be adjustable, for example, based on the current mission. In some cases, one or more of the first, second, and third thresholds may be set by the user. According to various implementations, at least one of the first, second, and third thresholds is automatically determined based on the current mission and / or one or more mission tasks.

[0020] FIG. 2 shows a more detailed flowchart for determining one or more corrective actions performed based on RUL according to the embodiments described herein. The system normality unit 201 monitors the normality of asset k, determines the presence of a failure, and estimates the remaining useful life RUL_k (205). If it is determined that RUL_k is very small, for example, RUL_k < t_critical (210), the system can protect itself by, for example, proceeding to a fail-safe mode (212). According to various implementations, the impact of the fail-safe mode on other operational goals can be evaluated, and a plan to return from the fail-safe mode can be made. Such a plan may involve maintenance measures for repairing and / or replacing the item, or other appropriate measures. The fail-safe mode may be configured to protect other mission objectives and trade off the functions of the failed component. The fail-safe measure releases the remaining useful life clock and allows more time to correct the problem. One or more fail-safe objectives 215 can be considered. These objectives may be pre-compiled to enable the fastest protection in case of a failure. If the failure is not repairable, this mission may be re-planned to account for the loss of function imposed by the failure of a particular component.

[0021] Although the remaining useful life is still small, but not as small as t_critical, for example, when it is determined that RUL_k < t_short (220), the system may be able to perform controller adaptation as a corrective measure. According to various embodiments, t_short is a parameter that can be adjusted based on operation constraints and mission constraints, and the system may be able to perform controller adaptation 222 as a corrective measure. Controller adaptation itself can be a complex task based on the complexity of the asset and based on the adaptation purpose 225 for a specific asset that is a control objective in the context of the controller. This can be formulated as an optimization problem affected by time constraints (i.e., RUL_k). The result of controller adaptation typically does not solve problems with the asset but results in an extension of the asset's life. Failures may still be addressed through maintenance, but the controller can gain additional time, for example, by reducing the load from a failed component.

[0022] Instead of controller adaptation, system reconfiguration can be used as a corrective measure. System reconfiguration involves using directly or indirectly redundant subsystems that at least partially restore operational capabilities. Reconfiguration can be achieved for various types of systems, including mechanical, electrical, or other types. For complex systems, finding partial and indirect redundancies is found by examining a large number of possible configurations and comparing the expected functions, and is formulated as an optimization problem affected by the adaptation purpose. In the context of system reconfiguration, the adaptation purpose is an objective function that includes the operational capabilities of the system.

[0023] Again, if the problem cannot be solved, this mission may be replanned to account for the loss of function of a specific component.

[0024] If it is determined that the remaining time is still large but not sufficient to enable cost-effective and safe maintenance (230) (i.e., for example, RUL_k < t_mission), the mission replanning module 232 can be triggered. This module 232 uses the mission objectives to evaluate the operational impairment considering the failed component. A new mission plan is evaluated to maximize the mission objectives 235 using the availability and readiness of new components. Such planning activities are an optimization task with numerous approaches, including various search algorithms and (especially) other task allocation algorithms. Such measures make it possible to continue operations in an optimal way (assuming a mix of available assets), while more permanent corrections may, in some cases, be pursued through maintenance and / or repair.

[0025] Finally, if it is determined that RUL_k is even larger such that RUL_k ≥ t_mission, but other means were taken due to a shorter RUL_k, the maintenance optimization module 242 is triggered. Here, the optimal maintenance measures are evaluated while also considering the logistics objectives 245. It may be determined whether the asset and / or asset component is repairable (250). If it is determined that the asset and / or asset component is not repairable (250), the mission replanning module 232 can be called. If it is determined that the asset and / or asset component is repairable (250), the asset and / or asset component may have maintenance 260 scheduled after the completion of the current mission.

[0026] Determining the optimal maintenance measures can be important and potentially traceable back to the manufacturing of the goods, but it also needs to be considered in relation to the needs of other assets, especially in fleet settings where two or more assets need to be operated. In such cases, many demands on logistics will compete for readily available assets and resources, and this problem can be interpreted as a large-scale dynamic multi-objective optimization problem. Here again, many approaches are available to solve this problem. The multi-objective nature of this problem means that multiple solutions are generated, which still require down-selection of the best solution. This is often done in conjunction with human operators who adopt contextual constraints that are not encoded in the final solution. Alternatively or additionally, this problem can be reduced to a single-objective problem by assigning weights to different objectives. In that case, a single solution can be found and pursued.

[0027] Figure 3 shows exemplary results of a simulation of a hypothetical asset with k=10 subsystems / components that fail at various times with different RUL_k values. As can be seen, depending on the remaining time of the component, the system's actions result in restoring the component's health, not necessarily immediately after the failure. For example, subsystem 1 310 has a RUL drop at time 312 and another RUL drop at time 314. Subsystem 1 310 is then repaired at time 316. Subsystem 7 320 has a significant RUL drop at time 322, prior to any RUL drop for subsystem 1 310. Subsystem 2 320 is not repaired until time 324, considerably later than subsystem 1 310 being repaired.

[0028] Figure 4 shows a more detailed diagram of each component, represented by RUL (blue) and operating status. For illustrative purposes, the operating status is zero when there is no failure. Note that components can still experience degradation that is considered normal. The operating status is "1" when the system must be in fail-safe mode. As shown in Table 1, it is "2" when the system is in controller adaptive mode, "3" when it is in mission replanning mode, and "4" when it is in maintenance mode.

[0029] [Table 1]

[0030] As can be seen, different situations require different measures that reflect policies and objectives for different modes and failures. The corrective actions taken may depend on the various objectives for different components, namely, fail-safe, control, mission, and / or logistical objectives.

[0031] The above method can be implemented in a computer using well-known computer processors, memory units, storage devices, computer software, and other components. A high-level block diagram of such a computer is shown in Figure 5. Computer 500 includes a processor 510 that controls the overall operation of computer 500 by executing computer program instructions that define such operation. The processor may use sensor information obtained from one or more sensors. It should be understood that the processor 510 may include any type of device capable of executing instructions. For example, the processor 510 may include one or more of the following: a central processing unit (CPU), a graphics processing unit (GPU), a field-programmable gate array (FPGA), and an application-specific integrated circuit (ASIC). The computer program instructions may be stored in a storage device 520 (e.g., a magnetic disk) and loaded into memory 530 when execution of the computer program instructions is desired. Thus, the steps of the method described herein may be defined by computer program instructions stored in memory 530 and controlled by the processor 510 that executes the computer program instructions. Depending on the implementation, the computer may perform the method steps as part of an on-premises server or a cloud-based service. The computer 500 may include one or more network interfaces 550 for communicating with other devices over a network. The computer 500 may also include other input / output devices (e.g., a display, keyboard, mouse, speaker, buttons, etc.) 560 that enable user interaction with the computer 500. Depending on the various embodiments, Figure 5 is a high-level representation of possible components of the computer for illustrative purposes, and the computer may include other components.

[0032] Unless otherwise indicated, all numbers used in this specification and the claims to represent shape, size, quantity, and physical properties should be understood in all cases to be modified by the term "approximately." Therefore, unless otherwise indicated, the numerical parameters described in the foregoing specification and the appended claims are approximations that may vary depending on the desired properties sought by those skilled in the art using the teachings disclosed herein. The use of numerical ranges by endpoints includes all numbers within that range (for example, 1 to 5 includes 1, 1.5, 2, 2.75, 3, 3.80, 4, and 5), as well as any range within that range.

[0033] The various embodiments described above may be implemented using circuits and / or software modules that interact to provide specific results. Those skilled in the field of computing technology can readily implement such described functions, either at the module level or as a whole, using knowledge commonly known in the art. For example, flowcharts illustrated herein may be used to generate computer-readable instructions / code for execution by a processor. Such instructions may be stored in a computer-readable medium and transferred to a processor for execution as known in the art. The structures and procedures shown above are merely representative examples of embodiments that can be used to facilitate the diagnosis of such inkjet ejectors.

[0034] The above description of exemplary embodiments is provided for illustrative and explanatory purposes only. It is not intended to be exhaustive of the concepts of the invention, or to limit the concepts of the invention to the exact forms disclosed. Many modifications and variations are possible in light of the above teachings. Any or all features of the disclosed embodiments can be applied individually or in any combination, and are not intended to be limiting, but are purely illustrative. The scope of the invention is intended to be limited not by the detailed description, but by the claims appended herein.

Claims

1. 1. A method comprising: monitoring the health of at least one asset; estimating a remaining useful life (RUL) of the at least one asset based on said monitoring; and categorizing the RULs of the assets into categories including those less than a time to complete a current mission and those greater than the time to complete the current mission; automatically performing one or more corrective actions during the current mission if the RUL is classified as being less than the time to complete the current mission, the one or more corrective actions being selected from among initiating a fail-safe mode, adapting a controller of the one or more assets, reconfiguring the system, and adjusting the current mission of the one or more assets; and if the RUL is classified as being equal to or greater than the time to complete the current mission, scheduling maintenance after the current mission of the at least one asset.

2. determining based on said monitoring whether a failure of said one or more assets has occurred; and The method of claim 1 , further comprising: classifying the RUL of the asset into a category based on the determination that a fault has occurred.

3. 2. The method of claim 1, wherein categorizing the RUL comprises categorizing the RUL of the asset into at least a first category, a second category, a third category, and a fourth category based on the remaining RUL.

4. Categorizing the RUL into categories includes: classifying the RUL of the at least one asset into the first category if the residual RUL is less than or equal to a first threshold; classifying the RUL of the at least one asset into the second category if the remaining RUL exceeds the first threshold and is less than or equal to a second threshold; classifying the RUL of the at least one asset into the third category if the residual RUL exceeds the second threshold and is less than or equal to a third threshold; and if a residual RUL exceeds the third threshold, classifying the RUL of the at least one asset into the fourth category.

5. The method of claim 4 , wherein at least one of the first threshold, the second threshold, and the third threshold is adjustable.

6. The method of claim 4 , wherein the third threshold is based on the time to complete the current mission.

7. initiating a fail-safe mode if the RUL is classified as being within the first category; adapting a controller for the one or more assets if the RUL is classified as being in the second category; and reconfiguring a system including the one or more assets if the RUL is classified as being within the second category; and adjusting the current mission of the one or more assets if the RUL is classified as being in the third category; and The method of claim 4 , further comprising: if the RUL is classified as being in the fourth category, scheduling maintenance after completion of the current mission.

8. The method of claim 7 , wherein adapting the controller of the one or more assets comprises adjusting one or more controller settings.

9. The method of claim 7 , wherein reconfiguring the one or more assets comprises one or more partial direct or indirect redundancies.

10. The method of claim 7 , wherein adjusting the current mission comprises one or more of replanning and rescheduling one or more mission tasks.

11. 1. A system comprising: A processor; and a memory storing computer program instructions which, when executed by the processor, cause the processor to: monitoring the health of at least one asset; estimating a remaining useful life (RUL) of the at least one asset based on said monitoring; and classifying the RULs of the assets into categories including those less than a time to complete a current mission and those greater than the time to complete the current mission; automatically performing one or more corrective actions during the current mission if the RUL is classified as being less than the time to complete the current mission, the one or more corrective actions being selected from among initiating a fail-safe mode, adapting a controller of the one or more assets, reconfiguring the system, and adjusting the current mission of the one or more assets; and if the RUL is classified as being equal to or greater than the time to complete the current mission, scheduling maintenance after the current mission of the at least one asset.

12. determining based on said monitoring whether a failure of said one or more assets has occurred; and The system of claim 11 , further comprising: classifying the RUL of the asset into a category based on the determination that a fault has occurred.

13. 12. The system of claim 11, wherein categorizing the RUL comprises categorizing the RUL of the asset into at least a first category, a second category, a third category, and a fourth category based on the remaining RUL.

14. Categorizing the RUL into categories includes: classifying the RUL of the at least one asset into the first category if the residual RUL is less than or equal to a first threshold; classifying the RUL of the at least one asset into the second category if the remaining RUL exceeds the first threshold and is less than or equal to a second threshold; classifying the RUL of the at least one asset into the third category if the residual RUL exceeds the second threshold and is less than or equal to a third threshold; and if a remaining RUL exceeds the third threshold, classifying the RUL of the at least one asset into the fourth category.

15. The system of claim 14 , wherein at least one of the first threshold, the second threshold, and the third threshold is adjustable.

16. The system of claim 14 , wherein the third threshold is based on the time to complete the current mission.

17. initiating a fail-safe mode if the RUL is classified as being within the first category; adapting a controller for the one or more assets if the RUL is classified as being in the second category; and reconfiguring the system using direct or indirect redundancy if the RUL falls into the second category; and adjusting the current mission of the one or more assets if the RUL is classified as being in the third category; and The system of claim 14 , further comprising: if the RUL is classified as being in the fourth category, scheduling maintenance after completion of the current mission.

18. The system of claim 17 , wherein adapting the controller of the one or more assets comprises adjusting one or more controller settings.

19. The system of claim 17 , wherein reconfiguring the system of the one or more assets includes one or more partial direct or indirect redundancies.

20. 20. The system of claim 17, wherein adjusting the current mission comprises one or more of replanning and rescheduling one or more mission tasks.

21. A non-transitory computer readable medium storing computer program instructions that, when executed by a processor, cause the processor to: monitoring the health of at least one asset; estimating a remaining useful life (RUL) of the at least one asset based on said monitoring; and classifying the RULs of the assets into categories including those less than a time to complete a current mission and those greater than the time to complete the current mission; automatically performing one or more corrective actions during the current mission if the RUL is classified as being less than the time to complete the current mission, the one or more corrective actions being selected from among initiating a fail-safe mode, adapting a controller of the one or more assets, reconfiguring the mission, and adjusting the current mission of the one or more assets; and if the RUL is classified as being equal to or greater than the time to complete the current mission, scheduling maintenance after the current mission of the at least one asset.

22. determining based on said monitoring whether a failure of said one or more assets has occurred; and 22. The non-transitory computer-readable medium of claim 21, further comprising: classifying the RUL of the asset into a category based on the determination that a fault has occurred.