Autonomous Vehicle Pickup Coordination on a Mapped Autonomy Grid
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicle (AV) operation on public roads is laborious due to the need for extensive mapping and labeling of ground truth data using sensors like LIDAR and stereoscopic cameras, which increases complexity and risk, especially in navigating dynamic environments.
Innovation Solution
An on-demand transport facilitation system that manages AV services by creating a 'autonomy grid' with detailed localization maps, allowing AVs to focus on specific road networks, and selects optimal vehicles (AVs or human-driven) based on proximity, cost, and time optimization, using a matching engine, cost optimizer, and rendezvous optimizer to coordinate pick-ups and transfers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If autonomous vehicles operate on public roads with extensive mapping and labeling of ground truth data, then localization precision is improved, but device complexity and risk increase
Solution Approach 1:
The system divides the service area into discrete grid cells with standardized mappings, breaking down the complex task of comprehensive road mapping into manageable segments. Each grid cell contains predefined localization data, allowing AVs to operate with high precision without requiring complete mapping of entire public road networks.
Solution Approach 2:
The system performs mapping and localization data preparation in advance for specific grid cells before AV operation. By pre-establishing the autonomy grid with ground truth data for selected areas, the system eliminates the need for real-time extensive mapping, reducing operational complexity while maintaining localization precision.
2Adaptability or versatility
If autonomous vehicles service the entire public road network, then adaptability is improved, but reliability decreases due to increased risk
Solution Approach 1:
The system segments the service area into an autonomy grid, allowing AVs to operate reliably within defined grid boundaries where mapping is complete and risks are managed. This segmentation enables gradual expansion of service coverage while maintaining high reliability within each grid cell.
Solution Approach 2:
The system applies different operational qualities to different spatial regions - full autonomous operation within mapped grid cells versus restricted or no operation in unmapped areas. This local quality approach allows the system to maximize reliability in serviced areas while progressively expanding adaptability as more grid cells are mapped.
3Productivity
If on-demand transport coordinates multiple vehicles for multi-hop rides, then productivity is improved, but device complexity increases
Solution Approach 1:
The system introduces a transport coordination service as an intermediary that manages multi-vehicle assignments and rider transfers. This mediator handles the complex coordination of multi-hop rides, matching riders with appropriate vehicle sequences, and managing transfers between vehicles, thereby improving transport productivity while containing coordination complexity within a dedicated system component.
Data Source
AI summary
An on-demand transport facilitation system can receive transport requests from requesting users throughout a given region, and select autonomous vehicles (AVs) and human driver to service the transport requests. The AV can operate on a mapped and labeled autonomy grid within the given region. For a given transport request, the transport system can determine an optimal pick-up location along the autonomy grid based on the current location of the requesting user and a current location of a selected AV, and transmit data indicating walking directions from the current location of the requesting user to the optimal pick-up location. The transport system may then coordinate the rendezvous by monitoring progress made by the requesting user and AV to the optimal pick-up location, and controlling the pace of the AV.