Presented at ITSC 2022

The study on “Cooperative path planning for multiple vehicles”, by Ayano Okoso et al. was selected for the presentation at 2022 IEEE International Conference on Intelligent Transportation Systems （ITSC）.

To maximally utilize the space within parking lots, investigations on high-density parking based on autonomous driving technology are in progress. It is necessary to plan not only efficient paths for entering and leaving but also paths for relocating vehicles obstructing entering/leaving at the appropriate timing and positions to achieve high-density parking. However, in conventional studies, paths could not be planned unless the destinations of each vehicle were set in advance. We proposed a method to optimize the destinations and paths of relocating vehicles simultaneously by abstracting a parking lot into a time-expanded network structure (two-dimensional grid graph and time axis) and by formulating a path planning problem as a multi-commodity flow problem on this network. This study enables optimal path planning for entering, leaving, and relocating under high-density parking conditions. This method is expected to be applicable to other path planning of various movable objects, such as robots in a factory.

Title: High Density Automated Valet Parking via Multi-agent Path Finding
Authors: Okoso, A., Otaki, K., Koide, S., Nishi, T.
Appears in: 2022 IEEE International Conference on Intelligent Transportation Systems
Presented: October 7, 2022
https://doi.org/10.1109/ITSC55140.2022.9922035