A5035286108- Robotic Path Planning Algorithms
- Reinforcement Learning in Robotics
- Robotic Locomotion and Control
- Robot Manipulation and Learning
- Distributed Control Multi-Agent Systems
- Advanced Neural Network Applications
- Robotics and Sensor-Based Localization
- Multi-Agent Systems and Negotiation
- Multimodal Machine Learning Applications
- Modular Robots and Swarm Intelligence
- Experimental Behavioral Economics Studies
- Domain Adaptation and Few-Shot Learning
- Evolutionary Game Theory and Cooperation
- Speech and dialogue systems
- Transportation and Mobility Innovations
- Neurogenetic and Muscular Disorders Research
- Adversarial Robustness in Machine Learning
Google (United States)
2019-2022
Long-range indoor navigation requires guiding robots with noisy sensors and controls through cluttered environments along paths that span a variety of buildings. We achieve this PRM-RL, hierarchical robot method in which reinforcement learning (RL) agents map to learn solve short-range obstacle avoidance tasks, then sampling-based planners where these can reliably navigate simulation; roadmaps are deployed on robots, them the shortest path likely succeed. In article, we use probabilistic...
Legged robots are physically capable of traversing a wide range challenging environments, but designing controllers that sufficiently robust to handle this diversity has been long-standing challenge in robotics. Reinforcement learning presents an appealing approach for automating the controller design process and able produce remarkably when trained suitable environments. However, it is difficult predict all likely conditions robot will encounter during deployment enumerate them at...
Joint attention - the ability to purposefully coordinate with another agent, and mutually attend same thing -- is a critical component of human social cognition. In this paper, we ask whether joint can be useful as mechanism for improving multi-agent coordination learning. We first develop deep reinforcement learning (RL) agents recurrent visual architecture. then train minimize difference between weights that they apply environment at each timestep, other agents. Our results show incentive...