filmov
tv
Learning Robust Control Policies for End-to-End Driving in Simulation | RA-L/ICRA 2020
Показать описание
This talk is streamed as part of a presentation for the IEEE International Conference on Robotics and Automation (ICRA) on the paper entitled:
Learning Robust Control Policies for End-to-End Autonomous Driving From Data-Driven Simulation.
Amini, A., Gilitschenski, I., Phillips, J., Moseyko, J., Banerjee, R., Karaman, S., & Rus, D. (2020). IEEE Robotics and Automation Letters, 5(2), 1143-1150.
Abstract:
In this work, we present a data-driven simulation and training engine capable of learning end-to-end autonomous vehicle control policies using only sparse rewards. By leveraging real, human-collected trajectories through an environment, we render novel training data that allows virtual agents to drive along a continuum of new local trajectories consistent with the road appearance and semantics, each with a different view of the scene. We demonstrate the ability of policies learned within our simulator to generalize to and navigate in previously unseen real-world roads, without access to any human control labels during training. Our results validate the learned policy onboard a full-scale autonomous vehicle, including in previously un-encountered scenarios, such as new roads and novel, complex, near-crash situations. Our methods are scalable, leverage reinforcement learning, and apply broadly to situations requiring effective perception and robust operation in the physical world.
Learning Robust Control Policies for End-to-End Autonomous Driving From Data-Driven Simulation.
Amini, A., Gilitschenski, I., Phillips, J., Moseyko, J., Banerjee, R., Karaman, S., & Rus, D. (2020). IEEE Robotics and Automation Letters, 5(2), 1143-1150.
Abstract:
In this work, we present a data-driven simulation and training engine capable of learning end-to-end autonomous vehicle control policies using only sparse rewards. By leveraging real, human-collected trajectories through an environment, we render novel training data that allows virtual agents to drive along a continuum of new local trajectories consistent with the road appearance and semantics, each with a different view of the scene. We demonstrate the ability of policies learned within our simulator to generalize to and navigate in previously unseen real-world roads, without access to any human control labels during training. Our results validate the learned policy onboard a full-scale autonomous vehicle, including in previously un-encountered scenarios, such as new roads and novel, complex, near-crash situations. Our methods are scalable, leverage reinforcement learning, and apply broadly to situations requiring effective perception and robust operation in the physical world.
0:09:59
Learning Robust Control Policies for End-to-End Driving in Simulation | RA-L/ICRA 2020
0:35:20
Learning Robust Driving Policies
0:29:44
Learning Robust Policies for Self-Driving
0:01:57
MIT Autonomous Vehicle: Learning Robust Sim-to-Real Control Policies
0:04:53
A Deep Reinforcement Learning Approach for Robust Control in Robotic Applications
0:01:01
Sim-to-(Multi)-Real: Transfer of Low-Level Robust Control Policies to Multiple Quadrotors
0:02:52
Learning Robust Autonomous Navigation and Locomotion for Wheeled-legged Robots
0:30:45
Deep Robust Reinforcement Learning and Regularization
0:47:39
2024 Risk Conference Replay: Leveraging MS-ISAC Services for Robust Multifamily Cybersecurity
0:57:59
Maryam Fazel, 'Policy Optimization for Learning Control Policies'
0:58:12
Webinar: Robust control strategies for musculoskeletal models using deep reinforcement learning
0:03:01
Learning and Deploying Robust Locomotion Policies with Minimal Dynamics Randomization
0:01:47
Learning Robust Manipulation Skills with Guided Policy Search
0:14:35
Multi-Agent Reinforcement Learning In Stochastic Games: From Alphago To Robust Control
1:04:10
Policy Optimization for Optimal Control with Guarantees of Robustness
0:56:02
Control Fundamentals
0:02:15
Using Data-Driven Domain Randomization to Transfer Robust Control Policies to Mobile Robots
0:04:56
Deep Adversarial Reinforcement Learning Method to Generate Control Policies Robust Against Worst Cas
0:28:07
Michael Zargham - Crafting Reliable Rules with Robust Control | PyData NYC 2023
0:02:29
Adaptively Robust Control Policy Synthesis Through Riemannian Motion Policies
0:03:00
Reinforcement Learning-Based Cascade Motion Policy Design for Robust 3D Bipedal Locomotion
0:26:03
Reinforcement Learning: Machine Learning Meets Control Theory
0:00:48
An optimal control approach to robust control of robot manipulators.
0:01:12
Optimal bounded robust control algorithm for quadcopter vehicles in presence of aggressive wind gust
Комментарии