Planning for Autonomous Driving via Interaction-Aware Probabilistic Action Policies
Devising planning algorithms for autonomous driving is non-trivial due to the presence of complex and uncertain interaction dynamics between road users. In this paper, we introduce a planning framework encompassing multiple action policies that are learned jointly from episodes of human-human intera...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2022-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9837917/ |
| _version_ | 1798040464864051200 |
|---|---|
| author | Salar Arbabi Davide Tavernini Saber Fallah Richard Bowden |
| author_facet | Salar Arbabi Davide Tavernini Saber Fallah Richard Bowden |
| author_sort | Salar Arbabi |
| collection | DOAJ |
| description | Devising planning algorithms for autonomous driving is non-trivial due to the presence of complex and uncertain interaction dynamics between road users. In this paper, we introduce a planning framework encompassing multiple action policies that are learned jointly from episodes of human-human interactions in naturalistic driving. The policy model is composed of encoder-decoder recurrent neural networks for modeling the sequential nature of interactions and mixture density networks for characterizing the probability distributions over driver actions. The model is used to simultaneously generate a finite set of context-dependent candidate plans for an autonomous car and to anticipate the probable future plans of human drivers. This is followed by an evaluation stage to select the plan with the highest expected utility for execution. Our approach leverages rapid sampling of action distributions in parallel on a graphic processing unit, offering fast computation even when modeling the interactions among multiple vehicles and over several time steps. We present ablation experiments and comparison with two existing baseline methods to highlight several design choices that we found to be essential to our model’s success. We test the proposed planning approach in a simulated highway driving environment, showing that by using the model, the autonomous car can plan actions that mimic the interactive behavior of humans. |
| first_indexed | 2024-04-11T22:07:58Z |
| format | Article |
| id | doaj.art-62c37cc0f0754d709df58395dc453288 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-04-11T22:07:58Z |
| publishDate | 2022-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-62c37cc0f0754d709df58395dc4532882022-12-22T04:00:38ZengIEEEIEEE Access2169-35362022-01-0110816998171210.1109/ACCESS.2022.31934929837917Planning for Autonomous Driving via Interaction-Aware Probabilistic Action PoliciesSalar Arbabi0https://orcid.org/0000-0002-9049-9161Davide Tavernini1https://orcid.org/0000-0001-5171-8803Saber Fallah2https://orcid.org/0000-0002-1298-1040Richard Bowden3https://orcid.org/0000-0003-3285-8020Centre for Automotive Engineering, University of Surrey, Guildford, U.K.Centre for Automotive Engineering, University of Surrey, Guildford, U.K.Centre for Automotive Engineering, University of Surrey, Guildford, U.K.Centre for Vision, Speech and Signal Processing, University of Surrey, Guildford, U.K.Devising planning algorithms for autonomous driving is non-trivial due to the presence of complex and uncertain interaction dynamics between road users. In this paper, we introduce a planning framework encompassing multiple action policies that are learned jointly from episodes of human-human interactions in naturalistic driving. The policy model is composed of encoder-decoder recurrent neural networks for modeling the sequential nature of interactions and mixture density networks for characterizing the probability distributions over driver actions. The model is used to simultaneously generate a finite set of context-dependent candidate plans for an autonomous car and to anticipate the probable future plans of human drivers. This is followed by an evaluation stage to select the plan with the highest expected utility for execution. Our approach leverages rapid sampling of action distributions in parallel on a graphic processing unit, offering fast computation even when modeling the interactions among multiple vehicles and over several time steps. We present ablation experiments and comparison with two existing baseline methods to highlight several design choices that we found to be essential to our model’s success. We test the proposed planning approach in a simulated highway driving environment, showing that by using the model, the autonomous car can plan actions that mimic the interactive behavior of humans.https://ieeexplore.ieee.org/document/9837917/Autonomous vehicleautonomous drivingdriver modelinghuman-robot interactioninteraction-aware motion prediction |
| spellingShingle | Salar Arbabi Davide Tavernini Saber Fallah Richard Bowden Planning for Autonomous Driving via Interaction-Aware Probabilistic Action Policies IEEE Access Autonomous vehicle autonomous driving driver modeling human-robot interaction interaction-aware motion prediction |
| title | Planning for Autonomous Driving via Interaction-Aware Probabilistic Action Policies |
| title_full | Planning for Autonomous Driving via Interaction-Aware Probabilistic Action Policies |
| title_fullStr | Planning for Autonomous Driving via Interaction-Aware Probabilistic Action Policies |
| title_full_unstemmed | Planning for Autonomous Driving via Interaction-Aware Probabilistic Action Policies |
| title_short | Planning for Autonomous Driving via Interaction-Aware Probabilistic Action Policies |
| title_sort | planning for autonomous driving via interaction aware probabilistic action policies |
| topic | Autonomous vehicle autonomous driving driver modeling human-robot interaction interaction-aware motion prediction |
| url | https://ieeexplore.ieee.org/document/9837917/ |
| work_keys_str_mv | AT salararbabi planningforautonomousdrivingviainteractionawareprobabilisticactionpolicies AT davidetavernini planningforautonomousdrivingviainteractionawareprobabilisticactionpolicies AT saberfallah planningforautonomousdrivingviainteractionawareprobabilisticactionpolicies AT richardbowden planningforautonomousdrivingviainteractionawareprobabilisticactionpolicies |