Self-optimizing on-vehicle network

Patent metadata

Field	Value
Publication number	US-12483293-B2
Application number	US-202418435363-A
Country	US
Kind code	B2
Filing date	Feb 7, 2024
Priority date	Feb 7, 2024
Publication date	Nov 25, 2025
Grant date	Nov 25, 2025

Abstract

Self-optimization of an ultra-wideband (UWB) network is performed. A configuration of UWB anchors of a vehicle is identified. A state vector is constructed based on the identified configuration, the state vector including values for status and ranging quality indicators for each of the UWB anchors. A next state is predicted based on the state vector using reinforcement learning to optimize UWB coverage, wherein a conditional probability of the next state depends only on the state vector. Unified configuration interface (UCI) commands are sent to the UWB anchors to reconfigure the UWB anchors based on the predicted next state.

First claim

What is claimed is:

1 . A method for performing self-optimization of an ultra-wideband (UWB) network, comprising:
identifying a configuration of UWB anchors of a vehicle; constructing a state vector based on the identified configuration, the state vector including values for status and ranging quality indicators for each of the UWB anchors; predicting a next state based on the state vector using reinforcement learning to optimize UWB coverage, wherein a conditional probability of the next state depends only on the state vector; and sending unified configuration interface (UCI) commands to the UWB anchors to reconfigure the UWB anchors based on the predicted next state.

2 . The method of claim 1 , wherein the configuration of the UWB anchors includes locations with respect to the vehicle and operational status of the UWB anchors.

3 . The method of claim 1 , wherein the next state is predicted using a Q-learning algorithm that learns an optimal policy maximizing expected value of a total reward over successive steps starting from a current state.

4 . The method of claim 1 , wherein the UCI commands are configured to cause the UWB anchors to one or more of increase radio frequency (RF) transmission power, reduce RF transmission power, change RF transmission frequency, assign a new time slot index, increase ranging time, or decrease ranging time.

5 . The method of claim 1 , wherein the configuration is an initial configuration of the vehicle as built.

6 . The method of claim 1 , wherein the configuration indicates one or more of the UWB anchors have become inoperative, and the self-optimization includes self-healing to command the remaining UWB anchors to adjust network parameters to reduce an effect of the inoperative UWB anchors.

7 . The method of claim 1 , wherein the configuration indicates one or more of the UWB anchors have been added, and the self-optimization includes self-configuration to command the UWB anchors to adjust network parameters to mitigate interference between neighboring UWB anchors.

8 . A system for performing self-optimization of an ultra-wideband (UWB) network, comprising:
one or more computing devices configured to:
identify a configuration of UWB anchors of a vehicle;
construct a state vector based on the identified configuration, the state vector including values for status and ranging quality indicators for each of the UWB anchors;
predict a next state based on the state vector using reinforcement learning to optimize UWB coverage, wherein a conditional probability of the next state depends only on the state vector; and
send unified configuration interface (UCI) commands to the UWB anchors to reconfigure the UWB anchors based on the predicted next state.

9 . The system of claim 8 , wherein the configuration of the UWB anchors includes the locations and operational status of the UWB anchors.

10 . The system of claim 8 , wherein the next state is predicted using a Q-learning algorithm that learns an optimal policy maximizing expected value of a total reward over successive steps starting from a current state.

11 . The system of claim 8 , wherein the UCI commands are configured to cause the UWB anchors to one or more of increase transmission power, reduce transmission power, change RF transmission frequency, assign a new time slot index, increase ranging time, or decrease ranging time.

12 . The system of claim 8 , wherein the configuration is an initial configuration of the vehicle as built.

13 . The system of claim 8 , wherein the configuration indicates one or more of the UWB anchors have become inoperative, and the self-optimization includes self-healing to command the remaining UWB anchors to adjust network parameters to reduce an effect of the inoperative UWB anchors.

14 . The system of claim 8 , wherein the configuration indicates one or more of the UWB anchors have been added, and the self-optimization includes self-configuration to command the UWB anchors to adjust network parameters to mitigate interference between neighboring UWB anchors.

15 . A non-transitory computer-readable medium comprising instructions for performing self-optimization of an ultra-wideband (UWB) network that, when executed by one or more computing devices, cause the one or more computing devices to perform operations including to:
identify a configuration of UWB anchors of a vehicle; construct a state vector based on the identified configuration, the state vector including values for status and ranging quality indicators for each of the UWB anchors; predict a next state based on the state vector using reinforcement learning to optimize UWB coverage, wherein a conditional probability of the next state depends only on the state vector; and send unified configuration interface (UCI) commands to the UWB anchors to reconfigure the UWB anchors based on the predicted next state.

16 . The non-transitory computer-readable medium of claim 15 , wherein the configuration of the UWB anchors includes the locations and operational status of the UWB anchors.

17 . The non-transitory computer-readable medium claim 15 , wherein the next state is predicted using a Q-algorithm that learns an optimal policy maximizing expected value of a total reward over successive steps starting from a current state.

18 . The non-transitory computer-readable medium of claim 15 , wherein the UCI commands are configured tocause the UWB anchors to one or more of increase transmission power, reduce transmission power, change RF transmission frequency, assign a new time slot index, increase ranging time, or decrease ranging time.

19 . The non-transitory computer-readable medium of claim 15 , wherein the configuration is an initial configuration of the vehicle as built.

20 . The non-transitory computer-readable medium of claim 15 , wherein one or more of:
the configuration indicates one or more of the UWB anchors have become inoperative, and the self-optimization includes self-healing to command the remaining UWB anchors to adjust network parameters to reduce an effect of the inoperative UWB anchors; and the configuration indicates one or more of the UWB anchors have been added, and the self-optimization includes self-configuration to command the UWB anchors to adjust network parameters to mitigate interference between neighboring UWB anchors.

Assignees

• Ford Global Tech Llc

Inventors

• Mutangana Jean De Dieu
• Vukovic Ivan
• Scheufler Nicholas Alexander
• Palakonda Sathyanarayana Chary
• Rollinger John Eric

Classifications

• H04B1/3822

  specially adapted for use in vehicles (H04B1/3827 takes precedence) · CPC title

• H04W24/02 Primary

  Arrangements for optimising operational condition · CPC title

• H04B1/719 Primary

  Interference-related aspects · CPC title