Multiverse Transformer: 1st Place Solution for Waymo Open Sim Agents Challenge 2023 (2306.11868v1)
Abstract: This technical report presents our 1st place solution for the Waymo Open Sim Agents Challenge (WOSAC) 2023. Our proposed MultiVerse Transformer for Agent simulation (MVTA) effectively leverages transformer-based motion prediction approaches, and is tailored for closed-loop simulation of agents. In order to produce simulations with a high degree of realism, we design novel training and sampling methods, and implement a receding horizon prediction mechanism. In addition, we introduce a variable-length history aggregation method to mitigate the compounding error that can arise during closed-loop autoregressive execution. On the WOSAC, our MVTA and its enhanced version MVTE reach a realism meta-metric of 0.5091 and 0.5168, respectively, outperforming all the other methods on the leaderboard.
- Pretr: Spatio-temporal non-autoregressive trajectory prediction transformer. arXiv:2203.09293, 2022.
- Simnet: Learning reactive self-driving simulations from real-world observations. In ICRA, 2021.
- Implicit latent variable model for scene-consistent motion forecasting. In ECCV, 2020.
- Large scale interactive motion forecasting for autonomous driving: The waymo open motion dataset. In ICCV, 2021.
- Trafficgen: Learning to generate diverse and realistic traffic scenarios. In ICRA, 2023.
- Vectornet: Encoding hd maps and agent dynamics from vectorized representation, 2020.
- Latent variable sequential set transformers for joint multi-agent motion prediction. In ICLR, 2022.
- Symphony: Learning realistic and diverse agents for autonomous driving simulation. In ICRA, 2022.
- Motiondiffuser: Controllable multi-agent motion prediction using diffusion. In CVPR, 2023.
- Jfp: Joint future prediction with interactive multi-agent modeling for autonomous driving. In PMLR, 2022.
- The waymo open sim agents challenge. arXiv:2305.12032, 2023.
- Wayformer: Motion forecasting via simple and efficient attention networks. arXiv:2207.05844, 2022.
- Scene transformer: A unified architecture for predicting multiple agent trajectories. In ICLR, 2022.
- Trace and pace: Controllable pedestrian animation via guided trajectory diffusion. In CVPR, 2023.
- Generating useful accident-prone driving scenarios via a learned traffic prior. In CVPR, 2022.
- Motion transformer with global intention localization and local movement refinement. In NeurIPS, 2022.
- Trafficsim: Learning to simulate realistic multi-agent behaviors. In CVPR, 2021.
- Bits: Bi-level imitation for traffic simulation. arXiv:2208.12403, 2022.
- Agentformer: Agent-aware transformers for socio-temporal multi-agent forecasting. In ICCV, 2021.
- Guided conditional diffusion for controllable traffic simulation. In ICRA, 2023.