Everyday Robots
So sad that this group worked on so many great ideas, and just shut down. That mobile manipulator did some really awesome things.
I think actually many of things Google Deepmind thinks about is scaling up robotics, which is quite important.
Papers that came out of it:
- Thinking While Moving Deep Reinforcement Learning with Concurrent Control (2020)
- ?
- AWOpt Learning Robotic Skills with Imitation and Reinforcement at Scale
- ?
- QTOpt Scalable Deep Reinforcement Learning for VisionBased Robotic Manipulation
- how they did RL for multi-task policy learning
- BCZ ZeroShot Task Generalization with Robotic Imitation Learning (2022)
- how they did imitation learning for multi-task policy learning
- Do As I Can Not As I Say Grounding Language in Robotic Affordances (2022)
- Deep RL at Scale Sorting Waste in Office Buildings with a Fleet of Mobile Manipulators (2023)
- Everyday robots shuts down here :(
- Inner Monologue Embodied Reasoning through Planning with Language Models (2023)
- AutoRT Embodied Foundation Models for Large Scale Orchestration of Robotic Agents
Robot foundation models
- RT1 Robotics Transformer for RealWorld Control at Scale
- RT2 VisionLanguageAction Models Transfer Web Knowledge to Robotic Control
Other