PFI-TT: Cloud-based Route Management Platform for Optimizing Last-Mile Logistics of Electric Truck and Drone Operations

University Of Missouri-Columbia, Columbia MO

Investigators

Abstract

The broader impact/commercial potential of this Partnerships for Innovation - Technology Translation (PFI-TT) project is to enable the safe and efficient use of unmanned aerial vehicles (UAVs or drones) and electric trucks for last-mile logistics, a multi-billion global market that accounts for 41% of the total supply cost and 8% of global greenhouse gas emissions. Specifically, the envisioned technology allows logistics service providers to establish reliable and optimized routing plans for a scalable fleet involving electric trucks and UAVs. The technology being developed in this project has the potential to: (i) increase the fleet utilization and productivity of last-mile service providers, (ii) reduce carbon footprint in supply chains, (iii) enhance operational safety and network communication reliability of drone-based deliveries, and (iv) decrease operational costs of last-mile logistics. The potential markets for the technology being developed in the project include many areas, such as medical delivery, e-commerce logistics, humanitarian operations, emergency response, and last-mile delivery. The methodological advancements and scientific understanding derived from this project can benefit policymakers, researchers and practitioners and strengthen the economic competitiveness of the United States. This project is developing and validating a cloud-based route optimization platform for electric trucks and UAVs to enable safe and efficient last-mile operations. A key technical challenge addressed in this project is the joint consideration of operational (i.e., optimizing vehicle’s route plan) and networking (packet forwarding strategy for communication among vehicles and ground servers) decisions for UAV and electric truck route management. The research objectives include the development and validation of (i) new optimization models and heuristic methods to efficiently solve the electric truck and UAV routing problem, (ii) new dynamic routing models that integrate machine learning and optimization for handling stochastic elements, while accounting for airspace and road network management, environmental factors and network communications, and (iii) network protocol design that features artificial intelligence (AI)-enabled algorithms for UAV trajectory planning based on awareness of environmental conditions. The project involves a modular strategy, thereby allowing the route planning algorithm to be scalable for any delivery methods (electric truck-only distribution, direct drone delivery, and hybrid truck-drone operations). The innovation in this project is evaluated based on field experimentation and pilot testing of a minimum viable product (MVP) in collaboration with multiple industrial partners. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.