Mapping Resolution via Image Clustering for Delivery Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In densely populated locations such as apartment complexes and office complexes, existing mapping technologies often provide low-resolution information, leading to difficulties in accurately identifying specific delivery locations, resulting in incorrect deliveries or misplaced items.
Innovation Solution
A system that utilizes a machine-learning model to generate high-resolution mapping information by analyzing images and associated location data from delivery agents, allowing for the augmentation of mapping databases with accurate latitude and longitude information, and continuously improves accuracy through agent feedback and image data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing mapping technologies are used in densely populated locations, then the mapping system can operate, but the resolution is insufficient to identify specific delivery locations accurately
Solution Approach 1:
The system performs preliminary actions by having agents capture images and location data at delivery locations during initial deliveries. This collected data is then processed to generate high-resolution mapping information before subsequent deliveries occur, enabling accurate location identification without requiring additional real-time searching
Solution Approach 2:
The system uses feedback from agents' image captures and location data to continuously improve and update the mapping database. Each delivery provides feedback that refines the mapping information, progressively enhancing resolution and accuracy for densely populated areas
2Productivity
If low-resolution mapping information is used, then the mapping system can function with existing data, but additional time is required for agents to locate and travel to the correct housing unit
Solution Approach 1:
High-resolution mapping information is prepared in advance through preliminary image capture and data collection during initial deliveries. This pre-generated detailed mapping data eliminates the need for agents to perform time-consuming location searches during subsequent deliveries, as the precise destination information is already available
3Reliability
If agents rely on low-resolution mapping information, then the system can operate with existing mapping data, but incorrect deliveries or misplaced items occur
Solution Approach 1:
The system continuously improves location identification accuracy by incorporating feedback from agents' actual delivery locations. Image data and location information from each delivery serve as feedback that refines the mapping database, progressively eliminating incorrect location identifications and ensuring accurate deliveries
Data Source
AI summary
In some examples, a system may receive over time, from one or more agent devices, and in association with a delivery location, a plurality of images and associated respective location data. Further, the respective location data associated with at least one of the images can differ from the respective location data associated with at least one other one of the images. The plurality of images are input to a machine-learning model that is trained to determine whether individual images include a threshold amount of information. Based at least on the machine-learning model indicating that the individual images satisfy the threshold amount of information, the system determines, based on at least one of averaging or clustering of the respective location information associated with the plurality of images, a consensus location for the delivery location. The system stores the consensus location information as mapping information associated with the delivery location.