Multimodal Delivery Routing for Inaccessible Last-Meter Drop-Offs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in delivering items to users due to obstacles such as poor road conditions, restricted access, and environmental factors like high winds or animals, which can hinder the use of unmanned aerial vehicles (UAVs) and delivery robots, leading to increased user effort and reduced delivery success rates.
Innovation Solution
The implementation of an AI-driven system that evaluates delivery feasibility and preference by analyzing data such as road conditions, weather, crime data, and user presence, using decision trees and machine learning algorithms to determine the best delivery method, whether through UAVs, delivery robots, or manual pickup, and adjusts delivery plans accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If autonomous vehicles are used for delivery, then delivery automation is improved, but delivery capability to inaccessible locations deteriorates
Solution Approach 1:
The delivery system is divided into two independent components: an autonomous vehicle for transportation and a separate delivery vehicle (UAV or delivery robot) for final delivery. The AV delivers items to a location near the user's residence, then a second delivery vehicle completes the delivery to the specific inaccessible location. This segmentation allows each component to specialize in its optimal function, resolving the contradiction between automation and adaptability.
Solution Approach 2:
A delivery vehicle acts as an intermediary between the autonomous vehicle and the final delivery location. The delivery vehicle (UAV or robot) receives items from the AV and transports them to locations that the AV cannot access directly, such as balconies, fenced backyards, or front doors. This intermediary resolves the limitation of AV accessibility while maintaining full automation.
2Adaptability or versatility
If delivery vehicles are used to reach inaccessible areas, then delivery capability is improved, but delivery success rate deteriorates due to environmental factors
Solution Approach 1:
The system uses sensor data, weather information, and environmental conditions to continuously monitor delivery feasibility. Before attempting delivery, the system evaluates factors such as high winds, animals, tree coverage, and crime data. Based on this feedback, the system dynamically adjusts delivery plans, selecting the most appropriate delivery vehicle type or timing to maximize success rate while maintaining capability to reach inaccessible locations.
Solution Approach 2:
The system dynamically selects between different delivery vehicle types (UAV vs. delivery robot) based on real-time environmental conditions. For example, UAVs are selected when weather conditions are favorable, while delivery robots are chosen when aerial delivery is risky. This dynamic adaptation allows the system to maintain high delivery success rates while preserving the ability to deliver to inaccessible areas.
3Adaptability or versatility
If users manually retrieve items from AVs, then delivery flexibility is improved, but user effort increases
Solution Approach 1:
The system implements automated delivery where the delivery vehicle autonomously completes the final delivery step without requiring user intervention. The AV drops off items at a location accessible to the delivery vehicle, and the delivery vehicle (UAV or robot) automatically transports items to the final destination. This self-service approach eliminates manual user effort while maintaining the flexibility to deliver to various locations.
Data Source
AI summary
Systems, methods, and computer-readable media are disclosed for automated multimodal delivery. Example methods may include determining information associated with the delivery; determining, based on at least a first portion of the information, that a first confidence level indicative of a delivery capability using a first vehicle is above a first threshold; and determining, based on at least a second portion of the information, a second confidence level indicative of a delivery preference for delivery using the first vehicle relative to a second vehicle.