ANN Drop-Off Spot Selection for Reliable Autonomous Delivery

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Package delivery systems rely on human judgment for selecting drop-off spots, which can be inefficient and prone to errors due to changing environments and frequent delivery demands, leading to obstructed or unreachable spots.

Innovation Solution

An artificial neural network (ANN) is trained to autonomously determine drop-off spots within a delivery destination based on previously designated spots, considering additional data such as object size, weather, and crime statistics, to select optimal locations with high confidence levels.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If human delivery personnel manually select drop-off spots, then delivery operations can be performed with simple systems, but the system is prone to errors due to changing environments and frequent delivery demands

Engineering Contradiction:
Improvedelivery success rateVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system enables autonomous vehicles to automatically select drop-off spots using trained neural networks, eliminating the need for human judgment in each delivery decision. The vehicle independently processes environmental data and makes delivery location decisions, achieving self-service operation that improves reliability while maintaining operational simplicity

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The neural network is trained in advance using historical delivery data and environmental information from multiple sources (aerial images, ground-level images, delivery logs). This preliminary training enables the system to make accurate real-time decisions without complex runtime processing, resolving the contradiction between reliability and complexity

Inventive Principle:
Principle #10Preliminary action

2Productivity

If human judgment is used for selecting drop-off spots, then the system can adapt to various environments, but it is inefficient and prone to errors due to changing environments and frequent delivery demands

Engineering Contradiction:
Improvedelivery efficiencyVSAvoiddelivery success rate
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The system replaces human mechanical judgment with an automated neural network that processes environmental data from multiple sources (aerial images, ground-level images, crime statistics, weather data). This substitution eliminates human limitations in processing speed and consistency, simultaneously improving both productivity and reliability through automated, data-driven decision-making

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system incorporates feedback mechanisms by analyzing historical delivery data, recipient feedback, and environmental changes to continuously improve drop-off spot selection. The neural network learns from past deliveries and adjusts its predictions, enabling efficient and reliable performance even in changing environments

Inventive Principle:
Principle #23Feedback

3Adaptability or versatility

If traditional methods are used to determine drop-off spots, then the system is simple to operate, but it cannot systematically consider factors that human delivery personnel might not consider

Engineering Contradiction:
Improveenvironmental adaptationVSAvoidoperational simplicity
Core Design Contradiction:
Adaptability or versatilityVSEase of operation

Solution Approach 1:

The neural network is designed to process multiple types of input data simultaneously (aerial images, ground-level images, crime statistics, weather data, delivery logs) and adapt to various delivery environments. This multi-functional capability enables the system to systematically consider diverse factors across different locations while maintaining a unified, simple operational interface through automated decision-making

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

Data Source

PatentEP3662335B1Model for determining drop-off spot at delivery location
Publication Date: 2024.06.12 WING AVIATION LLC
  • EP3662335B1 patent drawingFigure 1
  • EP3662335B1 patent drawingFigure 2
  • EP3662335B1 patent drawingFigure 3

AI summary

An example system includes a delivery vehicle, a sensor connected to the delivery vehicle, and a control system that determines a delivery destination for an object. The control system receives sensor data representing a physical environment of at least a portion of the delivery destination and determines a drop-off spot for the object within the delivery destination by way of an artificial neural network (ANN). The ANN is trained to determine the drop-off spot based on previously-designated drop-off spots within corresponding delivery destinations and includes an input node that receives the sensor data, hidden nodes connected to the input node, and an output node connected to the hidden nodes that provides data indicative of a location of the drop-off spot. The control system additionally causes the delivery vehicle to move to and place the object at the drop-off spot.