Autonomous Fleet Restricted-Area Mapping With UAV Sensor Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Managing fleets of autonomous vehicles to establish restricted areas is challenging due to the difficulty in obtaining precise location information for structures or features within a large geographic area, leading to potential operational restrictions that are either too broad or insufficiently protective.
Innovation Solution
A method and system for optimizing restricted areas by collecting environmental sensor data using UAVs to determine precise locations of structures or features within a general restricted area, allowing for the creation of tailored, minimally impactful specific restricted areas.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a fleet of autonomous vehicles is deployed to serve multiple geographic areas, then service coverage and accessibility are improved, but vehicle utilization efficiency deteriorates due to vehicles sitting idle during off-peak hours in specific zones
Solution Approach 1:
The system dynamically reassigns vehicles between geographic zones based on real-time and predictive demand data. Vehicles are not statically assigned to single zones but are dynamically routed to wherever demand is highest, allowing the fleet to adapt to changing utilization patterns across different geographic areas and time periods.
Solution Approach 2:
Vehicles serve multiple functions and multiple geographic zones rather than being dedicated to a single location. The same vehicle can serve Zone A during daytime hours and Zone B during evening hours, or be deployed to different locations based on predictive demand, making the fleet universally applicable across multiple service areas.
2Device complexity
If traditional dispatch methods are used without predictive analytics, then system complexity is reduced, but responsiveness to changing demand patterns deteriorates
Solution Approach 1:
The system performs preliminary actions by predicting future demand patterns and proactively positioning vehicles in advance before demand actually occurs. Rather than reactively responding to requests, the dispatch system uses machine learning models to forecast where vehicles will be needed and pre-deploys them to those locations, improving responsiveness without requiring complex real-time optimization.
Solution Approach 2:
The system implements feedback loops where actual ride data, utilization metrics, and demand patterns are continuously fed back into the machine learning models. This feedback mechanism allows the system to learn from past performance and continuously improve its predictive accuracy, enabling better responsiveness while maintaining manageable system complexity through data-driven optimization.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Techniques for optimizing a restricted area for autonomous vehicle operations is provided. A fleet management system receives a definition of a general restricted area. The fleet management system collects information associated with the general restricted area. The fleet management system determines a specific restricted area based on the definition of the general restricted area and the collected information. The fleet management system controls one or more autonomous vehicles based on the specific restricted area. In some embodiments, the collected information includes aerial imagery and/or other environmental sensor data.