Corpus ID: 224803036

SMARTS: Scalable Multi-Agent Reinforcement Learning Training School for Autonomous Driving

@article{Zhou2020SMARTSSM,
  title={SMARTS: Scalable Multi-Agent Reinforcement Learning Training School for Autonomous Driving},
  author={Ming Zhou and Jun Luo and Julian Villela and Yaodong Yang and David Rusu and Jiayu Miao and Weinan Zhang and Montgomery Alban and Iman Fadakar and Zheng Chen and Aurora Chongxi Huang and Ying Wen and Kimia Hassanzadeh and Daniel Graves and Dong Chen and Zhengbang Zhu and Nhat M. Nguyen and Mohamed ElSayed and Kun Shao and Sanjeevan Ahilan and Baokuan Zhang and Jiannan Wu and Zhengang Fu and Kasra Rezaee and Peyman Yadmellat and Mohsen Rohani and Nicolas Perez Nieves and Yihan Ni and Seyedershad Banijamali and Alexander Cowen Rivers and Zheng Tian and Daniel Palenicek and Haitham bou Ammar and Hongbo Zhang and Wulong Liu and Jianye Hao and Jintao Wang},
  journal={ArXiv},
  year={2020},
  volume={abs/2010.09776}
}
Multi-agent interaction is a fundamental aspect of autonomous driving in the real world. Despite more than a decade of research and development, the problem of how to competently interact with diverse road users in diverse scenarios remains largely unsolved. Learning methods have much to offer towards solving this problem. But they require a realistic multi-agent simulator that generates diverse and competent driving interactions. To meet this need, we develop a dedicated simulation platform… Expand
MetaDrive: Composing Diverse Driving Scenarios for Generalizable Reinforcement Learning
  • Quanyi Li, Zhenghao Peng, Zhenghai Xue, Qihang Zhang, Bolei Zhou
  • Computer Science
  • ArXiv
  • 2021
TLDR
This work develops a new driving simulation platform called MetaDrive, which is highly compositional, which can generate an infinite number of diverse driving scenarios from both the procedural generation and the real traffic data replay, and construct a variety of RL tasks and baselines in both single-agent and multi-agent settings. Expand
A Reinforcement Learning Benchmark for Autonomous Driving in Intersection Scenarios
TLDR
A benchmark for training and testing RL-based autonomous driving agents in complex intersection scenarios, which is called RL-CIS is proposed, and a set of baselines are deployed consists of various algorithms. Expand
Diverse Auto-Curriculum is Critical for Successful Real-World Multiagent Learning Systems
TLDR
It is argued that behavioural diversity is a pivotal, yet under-explored, component for real-world multiagent learning systems, and that significant work remains in understanding how to design a diversity-aware auto-curriculum. Expand
HAMMER: Multi-Level Coordination of Reinforcement Learning Agents via Learned Messaging
TLDR
This paper considers the case where there is a single, powerful, central agent that can observe the entire observation space, and there are multiple, low-powered, local agents that can only receive local observations and cannot communicate with each other. Expand
Towards a Very Large Scale Traffic Simulator for Multi-Agent Reinforcement Learning Testbeds
TLDR
A meso-macro traffic simulator that integrates mesoscopic and macroscopic traffic simulation model to improve efficiency and eliminate gridlocks and develops a graphic interface for users to visualize the simulation results in a web explorer. Expand
Learning to Build High-Fidelity and Robust Environment Models
  • Weinan Zhang, Zhengyu Yang, +5 authors Zhenguo Li
  • Computer Science
  • ECML/PKDD
  • 2021
TLDR
Experiments on continuous control scenarios demonstrate that the RL2S enabled methods outperform the others on learning high-fidelity simulators for evaluating, ranking and training various policies. Expand
LINDA: Multi-Agent Local Information Decomposition for Awareness of Teammates
TLDR
A novel framework, multi-agent Local INformation Decomposition for Awareness of teammates (LINDA), with which agents learn to decompose local information and build awareness for each teammate and significantly improves the learning performance, especially on challenging tasks. Expand
Deception in Social Learning: A Multi-Agent Reinforcement Learning Perspective
TLDR
This research review introduces the problem statement, defines key concepts, critically evaluates existing evidence and addresses open problems that should be addressed in future research. Expand
CARLA Real Traffic Scenarios - novel training ground and benchmark for autonomous driving
TLDR
This work introduces interactive traffic scenarios in the CARLA simulator, which are based on real-world traffic and are used to train agents using a reinforcement learning algorithm. Expand
Stepwise Goal-Driven Networks for Trajectory Prediction
TLDR
This paper argues that the goal of a moving agent may change over time, and modeling goals continuously provides more accurate and detailed information for future trajectory estimation, and presents a novel recurrent network for trajectory prediction, called Stepwise Goal-Driven Network (SGNet). Expand
...
1
2
3
...

References

SHOWING 1-10 OF 77 REFERENCES
Multi-Agent Connected Autonomous Driving using Deep Reinforcement Learning
  • P. Palanisamy
  • Computer Science, Mathematics
  • 2020 International Joint Conference on Neural Networks (IJCNN)
  • 2020
TLDR
This paper proposes the use of Partially Observable Markov Games (POSG) for formulating the connected autonomous driving problems with realistic assumptions and provides a taxonomy of multi-agent learning environments based on the nature of tasks, nature of agents and thenature of the environment to help in categorizing various autonomousdriving problems that can be addressed under the proposed formulation. Expand
MADRaS : Multi Agent Driving Simulator
TLDR
MADRaS provides a platform for constructing a wide variety of highway and track driving scenarios where multiple driving agents can train for motion planning tasks using reinforcement learning and other machine learning algorithms. Expand
Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environments
TLDR
An adaptation of actor-critic methods that considers action policies of other agents and is able to successfully learn policies that require complex multi-agent coordination is presented. Expand
The StarCraft Multi-Agent Challenge
TLDR
The StarCraft Multi-Agent Challenge (SMAC), based on the popular real-time strategy game StarCraft II, is proposed as a benchmark problem and an open-source deep multi-agent RL learning framework including state-of-the-art algorithms is opened. Expand
Safe, Multi-Agent, Reinforcement Learning for Autonomous Driving
TLDR
This paper applies deep reinforcement learning to the problem of forming long term driving strategies and shows how policy gradient iterations can be used without Markovian assumptions, and decomposes the problem into a composition of a Policy for Desires and trajectory planning with hard constraints. Expand
R-MADDPG for Partially Observable Environments and Limited Communication
TLDR
A deep recurrent multiagent actor-critic framework (R-MADDPG) for handling multiagent coordination under partial observable set-tings and limited communication and demonstrates that the resulting framework learns time dependencies for sharing missing observations, handling resource limitations, and developing different communication patterns among agents. Expand
MAVEN: Multi-Agent Variational Exploration
TLDR
A novel approach called MAVEN is proposed that hybridises value and policy-based methods by introducing a latent space for hierarchical control, which allows MAVEN to achieve committed, temporally extended exploration, which is key to solving complex multi-agent tasks. Expand
Grandmaster level in StarCraft II using multi-agent reinforcement learning
TLDR
The agent, AlphaStar, is evaluated, which uses a multi-agent reinforcement learning algorithm and has reached Grandmaster level, ranking among the top 0.2% of human players for the real-time strategy game StarCraft II. Expand
Networked Multi-Agent Reinforcement Learning in Continuous Spaces
TLDR
This paper proposes a fully decentralized actor-critic algorithm that only relies on neighbor-to-neighbor communications among agents in a networked multi-agent reinforcement learning setting, and adopts the newly proposed expected policy gradient to reduce the variance of the gradient estimate. Expand
Counterfactual Multi-Agent Policy Gradients
TLDR
A new multi-agent actor-critic method called counterfactual multi- agent (COMA) policy gradients, which uses a centralised critic to estimate the Q-function and decentralised actors to optimise the agents' policies. Expand
...
1
2
3
4
5
...