Vehicle Proximity Control Using ML-Based Portable Device Localization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle systems rely on physical keys for unlocking and actuating components, which can be inconvenient and prone to loss or unauthorized use, whereas modern systems aim to utilize portable devices like smartphones for proximity-based control without the need for physical keys.
Innovation Solution
A vehicle computer uses a machine learning program to determine the location of a portable device relative to the vehicle, predicting user actions and actuating components such as door locks based on the device's location, utilizing a weak supervised learning technique to refine its accuracy over time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If physical keys are used for unlocking and actuating vehicle components, then reliability of access control is maintained, but ease of operation deteriorates due to inconvenience and risk of loss
Solution Approach 1:
The patent replaces the mechanical key system with an electronic proximity-based system using portable devices. The vehicle computer detects the presence of a portable device through wireless communication signals and automatically actuates components such as door locks, eliminating the need for physical key insertion and mechanical interaction.
Solution Approach 2:
The patent introduces a vehicle computer as an intermediary between the user and vehicle components. The computer receives signals from portable devices, processes location information using machine learning programs, and automatically controls vehicle components based on predicted user actions, serving as a smart mediator that enhances both convenience and security.
2Ease of operation
If proximity-based control using portable devices is implemented, then ease of operation improves, but measurement precision of device location deteriorates
Solution Approach 1:
The patent implements a feedback mechanism where the machine learning program continuously refines its location predictions based on actual user actions. The system compares predicted actions with actual component actuations, uses this feedback to update the training dataset, and retrains the model to improve measurement precision over time.
Solution Approach 2:
The patent performs preliminary training of the machine learning program using a training dataset before actual operation. This preliminary action prepares the system to make accurate location predictions by pre-learning from labeled data, enabling the system to function effectively from the start while continuing to improve through ongoing feedback.
3Productivity
If machine learning programs are used to predict user actions, then productivity of vehicle control system improves, but device complexity increases
Solution Approach 1:
The patent implements a self-service mechanism where the machine learning program automatically trains and retrained itself using data collected from actual vehicle operations. The system collects operating data from actuated components, updates its training dataset, and retrains without external intervention, enabling the complex system to self-optimize and reduce operational complexity over time.
Data Source
AI summary
A computer includes a processor and a memory, the memory storing instructions executable by the processor to input a signal received from a portable device to a machine learning program trained to output a location of the portable device relative to a vehicle, collect operating data of one or more vehicle components, predict an action of a vehicle user based on the predicted location, and, based on the predicted action of the vehicle user, actuate one or more vehicle components. The machine learning program is trained with a training dataset that is updatable to include the signal, the output predicted location, the collected operating data, the predicted action, and an identified action performed by the vehicle user.