Autonomous Vehicle Control Release via Sensor Signal Processing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Autonomous vehicles face challenges in determining when to release user control efficiently, leading to potential delays or resource wastage, as the time required for users to access, exit, load, or unload varies significantly and is influenced by factors like traffic or physical disabilities.

Innovation Solution

The autonomous vehicle employs a computing system that processes sensor data from both interior and exterior sensors to identify release signals, determining a confidence score based on weighted signals to accurately assess when a vehicle service is complete, thereby releasing user control appropriately.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If the autonomous vehicle waits for a fixed duration to release user control, then the control release timing is simple to determine, but it leads to resource wastage and delays when users take longer than expected to access, exit, load, or unload

Engineering Contradiction:
Improvevehicle service efficiencyVSAvoiduser access/exit/loading/unloading time variability
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The system continuously monitors sensor data from interior and exterior sensors to detect user actions in real-time. This feedback mechanism allows the vehicle to determine when the user has completed access, exit, loading, or unloading operations, enabling dynamic control release timing that adapts to actual user behavior rather than relying on fixed durations.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The autonomous vehicle independently determines when to release user control by processing sensor data and identifying release signals without requiring manual input from the user. The system autonomously assesses whether the user has completed their actions and automatically transitions control back to autonomous operation.

Inventive Principle:
Principle #25Self-service

2Measurement precision

If the autonomous vehicle uses multiple sensors and complex processing to accurately determine user actions, then the control release timing precision is improved, but the system complexity increases

Engineering Contradiction:
Improveuser action detection accuracyVSAvoidsensor and processing system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system divides the monitoring task into distinct sensor zones and detection categories (interior sensors for user presence and actions, exterior sensors for contextual information). Each sensor type focuses on specific aspects of user behavior, and the processing system evaluates multiple independent signals to determine the overall release condition, improving accuracy through distributed detection.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The sensor system is designed to detect multiple types of user actions (access, exit, loading, unloading) using the same hardware infrastructure. The interior sensors and exterior sensors serve multiple detection purposes simultaneously, reducing the need for specialized sensors for each specific action type.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentUS20210018915A1Systems and Methods for Determining when to Release Control of an Autonomous Vehicle
Publication Date: 2021.01.21 UBER TECHNOLOGIES INC
  • US20210018915A1 patent drawing
  • US20210018915A1 patent drawing
  • US20210018915A1 patent drawing

AI summary

Systems and methods for controlling an autonomous vehicle are provided. In one example embodiment, a computer-implemented method includes providing a control of the autonomous vehicle to a user associated with a vehicle service. The method includes identifying one or more release signals indicative of one or more actions of the user. The method includes releasing the control of the autonomous vehicle from the user, based at least in part on the release signals.