Monte Carlo Tree Search and Knowledge Graphs for Decision Making in Autonomous Vehicles
- Delivery
- Available on this site
- Format
- Price
- Non-members (tax incl.):¥1,100 Members (tax incl.):¥880
- Publication code
- 20225254
- Paper/Info type
- Proceedings (Spring)
No.57-22
- Pages
- 1-7(Total 7 p)
- Date of publication
- May 2022
- Publisher
- JSAE
- Language
- English
- Event
- 2022 JSAE Annual Congress (Spring)
Detailed Information
Author(E) | 1) Irene Cara, 2) Mauro Comi, 3) Batrice Masini, 4) Beatrice Masini, 5) Ihsan Yalcinkaya, 6) Rutger Beekelaar |
---|---|
Affiliation(E) | 1) TNO, 2) TNO, 3) TNO, 4) TNO, 5) TNO, 6) TNO |
Abstract(E) | A fully autonomous vehicle must be able to drive in any condition and scenario. AI-based methods can increase the scalability of rule-based methods but they are not always as explainable and reliable as expert knowledge-based methods. The proposed approach investigates a novel method in the context of AI-based tactical decision making that combines Monte Carlo Tree Search, rules and an ontology-based Knowledge graph. The hybrid approach can increase scalability, trustworthiness and explainability of the framework. This decision making method is tested in two highway scenarios and one urban scenario in a simulation environment, proving the potential of the framework. |