Olfactive sensors and remediation systems

Abstract

Odor management for fleets of vehicles may be achieved using multivariate odor sensors comprising pluralities of reactive sites connected to transducers to detective physical changes in sites due to interactions with volatile organic compounds. Onboard systems may include multiple odor sensors, both in vehicle interior zones and outside the vehicle, to help determine the best odor remediation protocols to be applied in specific circumstances, and the efficacy of applied remediation efforts. Machine models for interpreting odor identities, intensities, and user preferences may be improved by harvesting fleet odor operations data and via querying users.

Description

[0001] OLFACTIVE SENSORS AND REMEDIATION SYSTEMS

[0002] TECHNICAL FIELD

[0003]

[0001] The invention relates to the field of olfactive systems such as in-vehicle olfaction systems (IVOS).

[0004] BACKGROUND

[0005]

[0002] More and more vehicles, notably automotive vehicles, are used in sharing models, either operated by traditional care renting companies or by car sharing companies that operate servers and smart phone terminals. The latter type of model is going to expand when autonomous vehicles will finally emerge. Companies that operate such markets want to ensure a user experience at least equivalent to that for a privately-owned vehicle. An odor left in a vehicle by a previous user is thus a problem that needs to be solved. Also, some users may violate the regulations that prohibit drug or alcoholic consumption when driving and the rental or car sharing agreements that forbid smoking or eating on board a car.

[0006]

[0003] Some of these problems have been partly addressed by providing a control system capable of setting on a heating, ventilation, and air conditioning (HVAC) system based on a detection by a sensor of carbon dioxide, sulfur dioxide or ammoniac or other type of odor in the cabin of a car. It is also known to include replaceable deodorizing filters connected to the vehicle HVAC system and to control the humidity level in the cabin with a moisture sensor to alert a vehicle fleet operator to have to replace the filters, thus preventing the degradation of the deodorization procedure. More recently, systems that are described as providing in-vehicle volatile organic compounds (VOC) detection have been advertised. Most odors can be traced to specific VOCs, and it is thus advantageous to be able to detect these VOCs to identify a specific odor.

[0007] SUMMARY

[0008]

[0004] The root cause of most odors is the presence of specific VOCs. Multivariate odor sensors may provide measurement data that enables a qualified quantification of a presence of a VOC or a mixture of VOCs in the indoor air of the vehicle and / or in the outdoor air, e.g., of a vehicle. This may be achieved, for example, using artificial intelligence (Al) models that are executed on-board the vehicle. Computer models may be maintained via communications with a server. Remedial measures to address odors may be determined and applied based on odors identified inside and outside the vehicle based on specific identities and intensities of the odors. Models for odor identification and remediation may be improved by collecting data from users and sensors regarding the odors and efficacy of attempted remediation efforts and / or in further view of weather, times, routes, cargos, air quality measures, and the like. Odor may be monitored at multiple locations within a passenger cabin, a cargo space, engine compartment, and / or outside the vehicle, etc. Associations may be made between the behaviors of drivers, routes taken, cargos carried, etc., to determine responsibility for incidents and / or derive strategies for preventive and / or remedial protocols based on sensed and / or anticipated odors based on routes, schedules, operators, cargos, and the like. Vehicles may be removed from service, delayed, serviced, and / or replaced based on odor events. Remedial protocols may include, for example, deploying deodorants, fragrances, solvents, etc., employing vehicle heating, ventilation, and air conditioning equipment, air filters, air purifiers, and the like.

[0009] BRIEF DESCRIPTION OF THE DRAWINGS

[0010]

[0005] Figure 1 is a block diagram of a system with central resources and a fleet of vehicles.

[0011]

[0006] Figure 2 is a block diagram of a vehicle onboard system.

[0012]

[0007] Figure 3 is a flow chart of example processes of an onboard system.

[0013]

[0008] Figure 4 is a flow chart of example processes for a fleet system.

[0014]

[0009] Figure 5 is a top view of an example passenger vehicle showing locations of air inlets and odor sensors.

[0015] DETAILED DESCRIPTION

[0016]

[0010] VOC sensing systems that deliver measurement data that match the high quality (repeatability, reproducibility, sensitivity, selectivity, specificity, as these terms are defined by a person of ordinary skill in the art of sensor quality), small form factor, small processing power footprint and low-cost criteria specified by the automotive industry is needed. The types of odors to be detected, as well as the types and behaviors of the drivers or passengers may vary from an area to the other, thus altering the recognition process and the determination of corrective actions required. A vehicle may be equipped both to autonomously assess its situation and act locally, and to receive and share information with central fleet operations and / or other vehicles and facilities.

[0017] [Oil] Figure 1 is a block diagram of an example system 1000 with central resources and a fleet of vehicles. The central resources include a geographic information system (GIS) 1008, other information sources 1010, a central facility 1004, and a maintenance facility 1006. In practice, the functionality of these resources may be arranged in a wide variety of ways. The GIS 1008 may include both global positioning satellite (GPS) resources and other beacons and information sources, such as cellular networks, which provide information about geographic positions, traffic, local conditions, and the like. The other information sources 1010 may include information feeds such as governmental announcements, web resources, and the like. The central facility 1004 may be an individual location housing a server or distributed computing capabilities which are used to coordinate vehicle fleet operations and maintenance. The maintenance facility 1006 may be a single vehicle service location, a group of service locations, or an affiliated or unaffiliated network work of service locations available for use in maintaining the fleet of vehicles 1002.

[0018]

[0012] In the example of Figure 1, the fleet of vehicles includes cars 1002 A, 1002B, and 1002C. In practice, these may be autonomous, semi-autonomous, and / or traditional vehicles. It will be appreciated that the techniques described herein for the management of olfactive issues in fleets of vehicles may be applied to groups of objects including any combination of vehicles, mobile equipment, mobile shelters, buildings, and / or rooms or other areas in buildings.

[0019]

[0013] Figure 2 is a block diagram of a vehicle onboard system 2000 that may be used in implementing system 1000 of Figure 1. Such a system may be built into a vehicle, for example, or provided as a standalone, e.g., after- market, option. The vehicle system 2000 includes a communications system 2004 that may be used to gather GIS and other information and to communicate with the central facility 1004 and maintenance facility 1006 of Figure 1. The communications system 2004 of Figure 2 is connected to a local computing device 2002 which coordinates the operations of the vehicle system 2000. The local computing device 2002 may be a central processing unit, microcontroller, computer or the like, or be embodied in distributed computing elements throughout a vehicle. The local computing device 2002 communicates with environmental equipment 2006, which may include onboard heating, ventilation, and air conditioning (HVAC) equipment, air filters, and / or air purifiers. The local computing device 2002 is also connected to other vehicle sensors, which may include, for instance, temperature, humidity, and / or occupancy sensors.

[0020]

[0014] The local computing device 2002 is also connected to olfactory sensors 2020 and / or olfactory effectors 2022. The olfactory sensors 2020 may detect odors within and around the vehicle, and the effectors 2022 may be used to remediate odors and / or calibrate the sensors.

[0021]

[0015] The olfactory sensors may be capable additionally or alternatively of detecting odorless Volatile Organic Compounds (VOCs). Odorless VOCs may be noxious when exposure is higher than a threshold. Herein the term “odor” generally refers to a condition that may be sensed by a multivariate sensor regardless of whether the condition may be sensed by the human olfactory system, e.g., either as odorless or beneath human sensory perception. The techniques described herein may be applied to the detection and / or remediation of emissions which may or may not be odorless to humans.

[0022]

[0016] Figure 3 is a flow chart of example operations 3000 that may be used in system 1000 of Figure 1 and system 2000 of Figure 2. At step 3002 of Figure 3, the onboard system 2000 of a vehicle is configured. Configuration may include receiving information from the GIS 1008, other information sources 1010, central facility 1004, and / or maintenance facility 1006 of Figure 1 , for example. Configuration data may also be built into the vehicle, provided by nearby by vehicles, and / or entered via a user interface of the vehicle, for example. Configuration information may include initial olfactory sensor calibration data and / or recommended odor avoidance and / or remediation techniques, such as sequences for the use of HVAC equipment and / or deodorants to remedy odors and / or routing information for avoiding odor sources. Sequences for the use of HVAC equipment and / or deodorants to remedy odors may include precautionary measures based on anticipated issues associate with cargos, drivers, occupants, users, routes, etc.

[0023]

[0017] At step 3004, the onboard system 2000 may receive and / or detect situational information such: geographic location, maps, traffic conditions, weather, air quality, and / or alerts; and / or information about drivers, riders, cargo, routes, destinations, and / or schedules.

[0018] At step 3006, sensor data is collected, and at step 3008 the data is analyzed. Notably, in step 3008, data for each sensor may be viewed in the context of other sensors and the situation of the vehicle, such that, for example, location, trip history, humidity, and temperature may be considered when interpreting olfactory sensor outputs.

[0019] At step 3010, user input may be solicited to aid in understanding the nature and severity of a detected odor. The users may be a driver or other occupant of the vehicle, a person in the vicinity of the vehicle, and / or a person responsible for the cargo or cargo operations. A user may be asked to identify the probable source of the odor, whether the odor is agreeable, and to what degree the level of the odor is acceptable.

[0024]

[0020] Users may be asked about an odor for any of several of reasons. For example, user input may be used to help in the initial identification of an odor. Additionally or alternatively, user input may be solicited after the system has identified an odor and its intensity. In either case, the system can use user input to determine user terminology for the odor and / or, their relative comfort or discomfort with the odor or its intensity, either to help with system learning or just confirm a determination already made by the system. By sampling responses from a variety of users, the system may learn, or refine, a variety of terms and preferences associated with various odors and intensity levels.

[0025]

[0021] User inputs may be simple responses, e.g., yes / no or ratings on a scale from 1 to ten. For example, once the system has achieved a desired certainty about a determination, e.g., where a probable accuracy exceeds a threshold, the system may ask the user to confirm a determination - or not. The response may then be used to adjust the system accordingly.

[0026]

[0022] At step 3012, the system may determine that remedial action is warranted to address an odor. Remedial actions may include, for example: instructions to a driver or occupant; activation of HVAC heating, venting, and / or cooling features; activation of an air purifier and / or air filter; scheduling maintenance; and / or deploying a deodorant, e.g., an aromatic surface, vapor, mist, or powder. In some circumstances, e.g., out of safety or legal concerns, such as with the detection of intoxicants such as drugs or alcohol, remedial may include preventing driving, stopping the vehicle, and / or reporting to local authorities.

[0027]

[0023] At step 3014, the system may check whether remedial measures have been effective. This may include taking additional sensor readings, checking on the status of effectors, and / or receiving confirmations from users contacted in step 3012.

[0028]

[0024] At step 3008, the onboard system 2000 may store and / or report results of operations to the central facility 1004 and / or maintenance facility 1006.

[0029]

[0025] It will be appreciated that the sequence of operations shown in Figure 3 is one of myriad ways in the operations described may be executed in turn or simultaneously.

[0026] Not shown in Figure 3 are additional options for operations such as procedures for clearing and / or recalibrating olfactory sensors and procedures for resetting operations and replacing and / or refilling system components and supplies.

[0030]

[0027] Figure 4 is a flow chart of example machine learning processes which may be used in fleet system 1000 of Figure 1, onboard system 2000 of Figure 2, and onboard operations 3000 of Figure 3. Step 4002 is the initial odor vector learning for the olfactory sensors and associated operations. This may include exposing sensors to a wide variety of chemicals and / or captured odors. This training may be done in the field and / or in a laboratory setting. Multivariate olfactory sensors, for example, can provide complex outputs reflecting plural measured aspects of air samples. Associating these outputs with specific odor sources can be complex and require repeated trials, experimentation, and training. This may involve fully automated machine learning, and / or human inputs. For examples, sensor data may be taken in a variety of situations, and the resulting sensor data may be annotated and / or cleaned by human technicians to assist in generating models for interpreting sensor data.

[0031]

[0028] Further, for purposes of interacting with various environments and users, it may be necessary to develop knowledge about local sources of odors and the language used by local inhabitants and others to describe those odors. Models of odors can be established before any vehicle is deployed and may be further refined using field data.

[0032]

[0029] At step 4004, information may be collected regarding odor remediation based on capabilities of vehicle HVAC and olfactory effectors vis-a-vis odor sources. Again, this information may be generated, e.g., via trials and machine learning, prior to deploying any vehicle, and then refined based on filed data, such as user input and / or sensor readings regarding the efficacy of remedial steps taken.

[0033]

[0030] At step 4006, knowledge of the relationships between cargos, passengers, drivers, locations, odors, times of day, seasons, weather conditions, tides, etc., can be developed in several ways. For example, an area may be mapped as to its geography and its industrial, commercial, agricultural, and residential uses on habits. Machine learning may be employed for an initial characterization of a region in which a system will be deployed. The model may be further refined using field data.

[0031] At step 4008, initial configurations and calibrations may be deployed to the fleet of vehicles, e.g., through direct communication with vehicles and / or via central or maintenance facilities.

[0034]

[0032] At step 4010, fleet odor data may be centrally collected. Notably, new information about odor sources and remediation may be used locally in individual vehicles and / or shared directly between vehicles. However, central gathering of experience may provide the most effective learning for the fleet.

[0035]

[0033] At step 4012, user input may be solicited regarding individual events and / or to refine understandings of odor nomenclatures and local preferences. At step 4014, maintenance facility personnel may provide more in-depth information about odor sources, issues with remediation techniques, and efficacy of attempted remediation.

[0036]

[0034] At step 4016, data accumulated in steps 4002 through 4014 may be collated, analyzed, and stored. At step 4018, the datasets collected may be used to further develop and refine methods for odor identification, avoidance, and / or remediation. The fleet may then be reconfigured accordingly.

[0037]

[0035] Figure 5 is a top view of an example passenger vehicle showing locations of air inlets and odor sensors. Not shown in Figure 5, alternatively or additionally a vehicle may have odor sensors and / or intake ports for odor sensors positioned at various other areas, such as a trunk, cargo hold, or engine compartment.

[0038]

[0036] The use of separate odor readings at various locations can help with identifying the source of the odor, who is responsible for the odor, and what to do about the odor.

[0039] Distinguishing the intensity of the odor at various locations in and / or around the vehicle assists in pinpointing the source and the best remedy.

[0040]

[0037] Sensing at multiple locations assists with the selection of a remedial protocol and determining the efficacy of such a protocol. For example, an external odor sensor may detect a noxious odor on the road prior to it being detected by an interior odor sensor or an occupant of the vehicle. The onboard system may react by closing the flow of exterior air into the cabin until the exterior air is clear again, without the occupants ever being aware or disturbed by the exterior odor. Similarly, during the mitigation of an interior odor, multiple sensors can be used to sense where the mitigation strategy is being effective or ineffective. This can be used for adjusting the actions of the mitigation and / or adjusting models for understanding which mitigating strategies for which odors are which locations.

[0041]

[0038] Responsibility can also be resolved through sensing at multiple locations. For example, consider the case of an aroma coming from spoiled food in the trunk of a rental car that was parked at the rental lot over a weekend. When a new user begins to drive, the foul odor is eventually noticed both by the user and the system. Rather than ascribing the odor to the new user, the system may have data about the food being in trunk of the car previously and be able to detect that the odor is coming from the trunk now, rather than the cabin where the new user entered. Therefore it may be inferred that the operation of the rental facility - who should have inspected the car between users - is responsible for the current odor.

[0042]

[0039] In contrast, consider a different case, where a new smell of tobacco smoke is sensed in a rental car, and further that there is no such smell outside the vehicle, and the smell is strongest near the driver, indicating that the driver is the source. Then, due to odor location information, it is inferable that the driver is liable for remediation of the smoking odor.

[0043]

[0040] Similarly, imagine a rental car being driven near a wastewater processing facility. The use of odor location information may be used to determine that the source is clearly from outside the car, and therefore not under the control of the driver, and not the fault or responsibility of the driver.

[0044]

[0041] Odor location information may also inform the choice of remediation protocol. For example, closing vents may be more appropriate for external odors. Location information may be combined with other information to determine a remediation protocol. For example, some odors may be acceptable to a first user of a rental vehicle but not to a second user. The first user may not appreciate remediation undertaken to address their body odor or perfume, but mitigation will be required before another user takes possession.

[0045]

[0042] The techniques described herein may be implemented in a processor configured to communicate with one or more multivariate odor sensors and a server, wherein each one of the one or more multivariate odor sensors has one or more core sensors, with core sensor having a plurality of reactive sites and one or more transducers. Each transducer is connected to a core sensor and configured to measure at least a physical property change induced by an interaction of a Volatile Organic Compound (VOC) sample characterizing the odor with the plurality of reactive sites. Each multivariate odor sensors is conditioned to receive at one or more air intake ports, each one connected to a core sensor and to receive one of indoor air from a passenger compartment of a vehicle and outdoor air from outside the vehicle, and configured to generate one or more of a first set of measurement data from indoor air and a second set of measurement data from outdoor air. One or more measurement data sets may be taken for each air intake port, allowing the processor to determine a qualified quantification of a presence in one or more of the passenger compartments of the vehicle and air outside the vehicle of one or more VOCs or odors selected in a list of one or more of VOCs or odors. The processor is further configured to determine one or more actions to be applied to at least one of the vehicle and a driver of the vehicle based at least on the said qualified quantification and one or more embedded VOC recognition models updated from time to time through a communication to the server.

[0046]

[0043] Many variations of a such a processor configuration are possible. For example, operations of the processor may be adapted for the case where an air intake is conditioned to receive one of indoor air from a passenger compartment of a vehicle and outdoor air from outside the vehicle.

[0047]

[0044] Similarly, a first set of measurement data and a second set of measurement data are both input to the one or more VOC detection models. The processor may be further configured to receive vehicle navigation data, wherein the second set of measurement data is input to the one or more VOC detection models together with actual and / or predicted qualified quantifications of an Air Quality Index (AQI) and / or VOCs in a navigation area of the vehicle. The actual and / or predicted qualified quantifications of AQI and / or VOCs in the navigation area of the vehicle take into account data received from the server, said data processed from actual and / or predicted qualified quantifications of AQI and / or VOCs received from other vehicles navigating or having navigated in the navigation area of the vehicle.

[0048]

[0045] Additionally or alternatively, the processor may be configured to execute embedded program instructions to control a vehicle HVAC system based on one of the actions determined to be applied to the vehicle, to control an odor dispenser based on one of the actions determined to be applied to the vehicle, and / or to send notifications to the server with a list of maintenance / cleaning actions to be executed at the point of arrival of the vehicle. The processor may be configured to send notifications to the server with an evaluation of the driver of the vehicle.

[0046] The techniques described herein may be embodied in a server configured to communicate with one or more such processors that are embedded on-board in one or more vehicles. Processing capabilities may be distributed between the processors and server in any number of ways. For example, the server may perform analyses for the interpretation of sensor data or assist the processors in doing so. Decisions as to maintenance, cleaning, remedial action, driver evaluations, and the like, may be made in a processor, a server, or via cooperation between a processor and a server.

[0049]

[0047] Many variations of such a server are possible. For example, the server may be configured to receive sets of measurement data from a plurality of vehicles. Similarly the server may collect data as to actions applied to a vehicle or a driver. The vehicles may be georeferenced.

[0050]

[0048] The server may be configured to use at least part of the data received from a plurality of vehicles to update a VOC recognition model and / or an action determination model in a supervised mode and an unsupervised mode. The server may be configured to process georeferenced AQI data.

[0051]

[0049] The server itself may be configured to determine a location of a source of VOC within the plurality of vehicles.

[0052]

[0050] Similarly, the server may be configured to receive, from a plurality of vehicles, a set of time and georeferenced data input by at least a passenger in the plurality of vehicles in relation to a context in a navigation area of each of the plurality of vehicles, the set of time and georeferenced data being used by the server in a learning procedure of at least one of a VOC recognition model and an action determination model. The server may be further configured to send to at least a vehicle in the plurality of vehicles a request to transmit a set of time and georeferenced data in relation to a context in a navigation area of the vehicle.

[0053]

[0051] The server may be further configured to receive a set of vehicle referenced data input by an operator in a maintenance station of the plurality of vehicles, the set of vehicle referenced data being used by the server in a learning procedure of at least one of a VOC recognition model and an action determination model. Similarly, the server may be configured to send to a maintenance station a request to transmit a set of vehicle referenced data.

[0054]

[0052] For purposes of illustration, the techniques herein have been described in terms of vehicle operations, but they may also be applied to other fixed or moveable spaces, such as pieces of equipment, cargo containers, cargo spaces, residential or commercial space, etc., where the adaptation of processors, servers, and other item illustrated in Figs. 1-5 may be useful in the detection, identification, mitigation, and rectification of odors.

Claims

CLAIMSWhat is claimed is:

1. An onboard system for odor management in a vehicle, comprising: one or more odor sensors, each odor sensor being a multivariate odor sensor comprising a plurality of reactive sites and one or more transducers connected to the plurality of reactive sites, whereby each transducer detects a physical property change induced by an interaction of one or more Volatile Organic Compounds (VOCs) with one or more of the plurality of reactive sites; a first air intake port connecting an interior point of a cabin of the vehicle to a first of the multivariate odor sensors; one or more computing devices, wherein the computing devices are configured to: analyze multivariate odor sensor data from the transducers; determine, from the multivariate odor data, an identity of a first odor and a first intensity level of the first odor within the cabin; determine, based on the first odor and the first intensity level, whether remedial action is warranted; and if remedial action is warranted, select a remedial action based on the first odor and the first intensity level and initiate the selected remedial action.

2. The onboard system of claim 1, further comprising one or more olfactory outputs, each olfactory output comprising a reservoir of aromatic material and a reservoir output valve, wherein the computing devices are configured to, if remedial action is warranted, to initiate the selected remedial action by activating one or more of by a heating, venting, and air conditioning (HVAC) operation and actuating one of the reservoir output valves.

3. The onboard system of claim 2, further comprising a second air intake port connecting a point of the vehicle exterior to the cabin to a second of the odor sensors, wherein the computing devices are configured to: determine, from the multivariate odor data from the first odor sensor and the second odor sensor, the identity of the first odor, the first intensity level of the first odor within the cabin, and a second intensity level of the first odor exterior to the cabin;determine, based on the first odor, and the first intensity level, and the second intensity level, whether remedial action is warranted; and if remedial action is warranted, select the remedial action based on the first odor, the first intensity level, and the second intensity level, and initiate the selected remedial action.

4. The onboard system of claim 3, wherein the computing devices are further configured to: query an occupant of the vehicle, via a user interface, as to whether the first odor is unpleasant; and adjust, based on a first response from the occupant a remedial, protocol for the first odor.

5. The onboard system of claim 4, wherein the computing devices are further configured to: query the occupant of the vehicle, via the user interface, as to whether the first intensity level of the first odor is unpleasant; and adjust, based on a second response from the occupant, the remedial protocol for the first odor.

6. The onboard system of claim 5, wherein the computing devices are further configured to: query the occupant of the vehicle, via the user interface, as to the efficacy of a first attempted remediation of the first odor; and record, based on a third response from the occupant, a user satisfaction with the first attempted remediation.

7. The onboard system of claim 6, wherein the computing devices are further configured to: query the occupant of the vehicle, via the user interface, to identify the first odor; and record, based on a fourth response from the occupant, a name for the first odor.

8. The onboard system of claim 7, wherein the computing devices are further configured to report, to a central facility, one or more of odors and intensities detected, remedial operations attempted, efficacies of attempted remediations, and usernames for detected odors.

9. The onboard system of claim 8, wherein the computing devices are further configured to receive, from the central facility, a revised protocol for remediation of an odor.

10. The onboard system of claim 9, wherein the computing devices are further configured to receive, from the central facility, a dispatch directing the vehicle to a maintenance facility.

11. The onboard system of claim 10, wherein the computing devices are further configured to: receive vehicle navigation data; and report, to the central facility, the time and location when odors and intensities are detected.

12. The onboard system of claim 11, wherein the computing devices are further configured to: receive, from the central facility, a map indicating where various VOCs are more likely to be detected; use the map in determining the identity of an odor.

13. The onboard system of claim 1, wherein the pluralities of reactive sites each comprise a plurality of cross-referenced oligopeptides.

14. The onboard system of claim 11, wherein each of the pluralities of reactive sites is grafted on a substrate with a thickness equal to or lower than 15 nanometers.

15. The onboard system of claim 12, wherein one or more of the transducers comprise one or more Mach-Zehnder Interferometers (MZIs).

16. A system for fleet odor management, the system comprising a central computing facility, the system being configured to: provide, to onboard systems deployed in a fleet of vehicles, configuration data for use in the interpretation sensorgrams provided by multivariate odor sensors each comprising one ortransducers responding to one or more physical changes in one or more reactive sites in a plurality of reactive sites in each multivariate odor sensor, the physical changes being induced by an interaction of one or more Volatile Organic Compounds (VOCs) plurality of reactive sites, whereby the configuration data allows the identification of odors and detection of odor intensity levels for odors present at one or more multivariate odor sensors associated with each vehicle; provide, to the onboard systems, remediation protocols for vehicle responses for detected odors and detected odor intensity levels; and collect, from the onboard systems, history of detected odors, detected odor intensities, attempted odor remediations, and results of attempted remediations.

17. The system of claim 16, wherein the system is further configured to collect user satisfaction information from occupants of a vehicle where a remediation has been attempted.

18. The system of claim 16, wherein the system is further configured to remove from service a vehicle where a remediation attempt has failed.

19. The system of claim 16, wherein the remediation protocols comprise the release of a deodorant.

20. The system of claim 19, wherein the remediation protocols comprise the release of a deodorant tailored to a detected odor.

Images