Autonomous Vehicle Dispatch Modes for Fleet-Level Objectives
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional transportation management systems struggle to optimize the use of autonomous vehicles beyond basic driving tasks, failing to manage fleet-level objectives such as vehicle maintenance, distribution, and service center utilization effectively.
Innovation Solution
A transportation management system that integrates fleet-level management architecture to instruct autonomous vehicles to fulfill ride requests while advancing objectives like maintenance, redistribution, and service center optimization, using strategies like no-passenger, destination, and incentive modes to minimize downtime.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If autonomous vehicles are dispatched to fulfill ride requests, then transportation service is provided, but fleet-level objectives such as maintenance, distribution, and service center utilization are not optimized
Solution Approach 1:
The system enables autonomous vehicles to perform multiple functions: fulfilling ride requests (transportation service) and advancing fleet-level objectives (maintenance, redistribution, service center optimization). The dispatch system integrates both individual vehicle tasks and fleet-wide goals into a unified management framework, allowing vehicles to transition between different operational modes based on system needs.
2Adaptability or versatility
If autonomous vehicles spend time on fleet-level tasks like maintenance and redistribution, then fleet optimization is improved, but vehicle utilization and productivity decrease
Solution Approach 1:
The system performs fleet-level tasks in advance or during low-demand periods. Vehicles are dispatched to service centers for maintenance or redistribution to optimal locations before they are needed, rather than taking them offline when demand is high. This preliminary action ensures fleet readiness while minimizing impact on current productivity.
Solution Approach 2:
The dispatch system dynamically adjusts vehicle allocation based on real-time conditions. When demand is low, more vehicles can be allocated to fleet-level tasks; when demand is high, more vehicles are dedicated to ride requests. This dynamic balancing allows the system to optimize both fleet-level objectives and vehicle utilization without fixed trade-offs.
3Reliability
If autonomous vehicles are taken offline for maintenance and redistribution, then fleet reliability is improved, but service availability and passenger experience deteriorate
Solution Approach 1:
The system uses autonomous vehicles to service themselves by dispatching them to service centers for maintenance and redistribution without human intervention. This self-service capability allows maintenance to be performed systematically across the fleet while minimizing disruption to service availability, as vehicles can be rotated through maintenance cycles independently.
Solution Approach 2:
The dispatch system continuously monitors fleet status, vehicle conditions, and service demand to make informed decisions about when to take vehicles offline for maintenance. This feedback loop ensures that maintenance scheduling does not significantly impact service availability, as the system can adjust dispatch patterns based on real-time conditions.
Data Source
AI summary
In one embodiment, a method includes one or more computing systems determining a fleet-level objective for a vehicle. The fleet-level objective is associated with instructing the vehicle to travel a route according to route criteria based on the fleet-level objective. The one or more computing systems may determine one or more of an urgency score indicative of an urgency to fulfill the fleet-level objective or a risk factor score indicative of an overall condition of the vehicle. Based on one or more of the urgency score or the risk factor score, the one or more computing systems may select a particular operating mode from a plurality of operating modes for the vehicle to operate. The one or more computing system may instruct the vehicle to operate in the particular operating mode so as to fulfill the fleet-level objective.