Pickup Vehicle Gesture Matching for Crowd Passenger Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies struggle to accurately identify an individual within a group of people from a distance, particularly in adverse conditions, using facial recognition or gesture-based systems, which are prone to errors and data privacy concerns.
Innovation Solution
A method involving a pick-up person and a target person equipped with electronic devices that communicate to guide the pick-up person to the target location, where the target person performs a waving gesture detected by environmental sensors, with characteristic movement data analyzed to identify the correct individual using augmented reality overlays.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If facial recognition is used to identify the target person, then identification can be performed, but it only works if the person is clearly visible and image data has been transmitted beforehand, making reliable recognition difficult at greater distances and raising data protection concerns
Solution Approach 1:
The patent replaces optical recognition systems (cameras and image processing) with a mechanical sensor system (accelerometers and gyroscopes) that detects physical movement patterns. This substitution enables identification at greater distances without requiring clear visual data transmission, as the sensors detect gesture mechanics rather than visual appearance.
Solution Approach 2:
The patent introduces movement pattern data as an intermediary between the target person and the pick-up person. Instead of directly transmitting visual identity data or requiring face-to-face recognition, the system uses characteristic movement patterns as a mediator that can be detected from a distance and processed to identify the target person.
2Ease of operation
If the traditional solution of giving the passenger the taxi's license plate number and the driver the name of the person being picked up is used, then identification can be performed, but it is prone to errors if the name is misspelled or is a common name
Solution Approach 1:
The patent replaces text-based identification (names and license plate numbers) with gesture-based identification using motion sensors. This substitution eliminates spelling errors and confusion from common names, as the system identifies persons through unique physical movement patterns detected by accelerometers and gyroscopes rather than through text input.
3Loss of information
If a larger sign visibly displayed behind the vehicle's windshield is used in the traditional solution, then the passenger can be identified, but the driver cannot identify the passenger this way
Solution Approach 1:
The patent transitions from two-way visual identification (signs visible to driver) to one-way sensor detection (gestures detected by vehicle sensors). This dimensional change allows the driver to identify the passenger through sensor data while the passenger identifies the driver through existing visual cues, enabling mutual identification without requiring large visible signs.
4Productivity
If gesture-based systems are used to identify the target person, then identification can be performed from a distance, but several people may be making the same gesture, presenting difficulties in identifying the correct individual
Solution Approach 1:
The patent applies local quality by analyzing specific characteristics of movement patterns rather than just the general gesture type. Instead of detecting only that a waving gesture occurred, the system analyzes the unique temporal and spatial characteristics of each person's waving pattern, enabling differentiation between multiple people performing the same gesture type.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention relates to a method for route guidance to a target person, in which a pick-up vehicle is guided to the target person (12), wherein the pick-up vehicle and the target person (12) are provided with electronic devices (200, 122) which can communicate with one another. A pick-up request is sent by the electronic device (122) of the target person (12), which is either transmitted directly to the pick-up vehicle or forwarded to the pick-up vehicle indirectly via a central server of a pick-up service. A navigation route to the pick-up location of the target person (12) is calculated by the electronic device (200) of the pick-up vehicle. When approaching the pick-up location, a prompt is sent by the electronic device of the pick-up vehicle (200) to the electronic device (122) of the target person (12), with which the target person (12) is prompted to perform a gesture. The movement data for the performed gesture are detected by the electronic device (122) of the target person (12) and transmitted to the electronic device (200) of the pick-up vehicle. The data from at least one environment observation sensor (150) is evaluated, by the electronic device (200) of the pick-up vehicle, for typical movement patterns corresponding to the performance of the gesture. With multiple detections of typical movement patterns of this gesture, the individual detected gestures are compared with the transmitted movement data and the detected gesture with the least deviation from the transmitted movement pattern is assigned to the target person (12).