# US 11,897,314 B2 · Method, computer program and system for improving a user experience > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,897,314 B2 - Document ID: US11897314B2 - Status: Granted - Kind: B2 - Application number: 17/153,007 - Filed: 2021-01-20 - Published: 2022-07-21 - Granted: 2024-02-13 - Earliest claimed: 2021-01-20 - Assignees: Bayerische Motoren Werke Aktiengesellschaft - Applicants: Bayerische Motoren Werke Aktiengesellschaft - Inventors: Yang Cao; Jilei Tian - Web page: https://patentgrep.com/patent/US11897314B2 ## CPC - B60H 1/00878: {the components being temperature regulating devices} - B60W 50/08: Interaction between the driver and the control system - G06N 7/01: Probabilistic graphical models, e.g. probabilistic networks - B60H 2001/00733: {Computational models modifying user-set values} - B60H: ARRANGEMENTS OF HEATING, COOLING, VENTILATING OR OTHER AIR-TREATING DEVICES SPECIALLY ADAPTED FOR PASSENGER OR GOODS SPACES OF VEHICLES - B60W: CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract Embodiments relate to a method, computer program and system for improving a user experience. The computer-implemented method for improving a user experience inside a vehicle comprises determining an environmental context inside the vehicle, wherein the environmental context is assigned a plurality of actions, which influences the environmental context. Further, the method comprises choosing at least one action of the plurality of actions for improving the user experience. ## Claims ### Claim 1 A computer-implemented method for optimizing a user experience inside a vehicle, comprising: receiving, by a receiver, a signal carrying information regarding an environmental context inside the vehicle from a sensor via a wireless or wired channel, wherein the sensor is either a device installed in the vehicle or a device included in a mobile electronic device of a user, wherein the environmental context is assigned a plurality of actions, which influences the environmental context; sampling, by a machine learning model, a reward for each action of the plurality of actions based on the environmental context and parameters for the machine learning model for sampling rewards for the plurality of actions; choosing, by the machine learning model, at least one action of the plurality of actions for optimizing the user experience based on the sampled rewards; controlling a vehicle system in the vehicle to implement the at least one chosen action to change the environmental context inside the vehicle; observing a reaction of the user in response to controlling the vehicle system to change the environmental context inside the vehicle; and updating the parameters for the machine learning model for sampling the rewards based on the reaction. ### Claim 2 (depends on 1) The method according to claim 1, wherein the at least one chosen action is assigned a plurality of arms, wherein each arm of the plurality of arms has a different influence on the environmental context; and further comprising: choosing an arm for optimizing the user experience; and controlling the vehicle system in the vehicle to implement the chosen arm to change the environmental context inside the vehicle. ### Claim 3 (depends on 2) The method according to claim 2, further comprising: assuming an initial arm parameter from a distribution of parameters for each arm of the plurality of arms; sampling an arm reward for each arm of the plurality of arms, wherein the environmental context and the initial arm parameters are used as input parameters for sampling the arm reward; and wherein choosing an arm of the plurality of arms is done in dependence of the sampled arm rewards. ### Claim 4 (depends on 3) The method according to claim 3, further comprising: adding information about the determined arm and the observed reaction to the input parameters for sampling the rewards. ### Claim 5 (depends on 4) The method according to claim 4, wherein the reaction of the user is to change the chosen action and/or arm; and further comprising setting the reward for the chosen action and/or arm to a minimum value. ### Claim 6 (depends on 4) The method according to claim 4, wherein the reaction of the user is no change of the chosen action and/or arm; and further comprising setting the reward for the chosen action and/or arm to a maximum value. ### Claim 7 (depends on 1) The method according to claim 1, wherein for sampling the reward a Gaussian noise and/or a Gaussian distribution is used. ### Claim 8 (depends on 1) The method according to claim 1, wherein for sampling the reward a linear Bayesian regression is used. ### Claim 9 (depends on 1) The method according to claim 1, wherein adding the information about the observed reaction changes at least one parameter of a used Gaussian noise and/or Gaussian distribution and/or linear Bayesian regression. ### Claim 10 (depends on 1) A computer program product comprising a non-transitory, computer-readable storage medium having program instructions embodied therewith, the program instructions executable by a device to cause the device to perform the method of claim 1. ### Claim 11 (depends on 1) A device comprising: a memory; and at least one processor operably coupled to the memory; and wherein the device is configured to perform the method of claim 1. ### Claim 12 (depends on 11) The device according to claim 11, wherein the device is portable by the user. ### Claim 13 (depends on 11) A vehicle comprising the device of claim 11.