Faraday Future would like you to purchase an $89,900 robot . As part of its latest revamp, the embattled electric car company is now pitching a lineup of robots, including humanoids, quadrupeds and a robotic arm. If that name doesn't ring any bells, that could be because the company has been going through a bit of a pivot over the past year in an attempt to salvage its bottom line, if not its reputation. That's become something of a theme for Faraday. The business generated a fair bit of hype years before it showed off its first production-ready electric car at CES 2017.

This paper introduces the Real-DRL framework for safety-critical autonomous systems, enabling runtime learning of a deep reinforcement learning (DRL) agent to develop safe and high-performance action policies in real plants (i.e., real physical systems to be controlled), while prioritizing safety! The Real-DRL consists of three interactive components: a DRL-Student, a PHY-Teacher, and a Trigger. The DRL-Student is a DRL agent that innovates in the dual self-learning and teaching-to-learn paradigm and the real-time safety-informed batch sampling. On the other hand, PHY-Teacher is a physics-model-based design of action policies that focuses solely on safety-critical functions. PHY-Teacher is novel in its realtime patch for two key missions: i) fostering the teaching-to-learn paradigm for DRL-Student and ii) backing up the safety of real plants. The Trigger manages the interaction between the DRL-Student and the PHY-Teacher. Powered by the three interactive components, the Real-DRL can effectively address safety challenges that arise from the unknown unknowns and the Sim2Real gap. Additionally, Real-DRL notably features i) assured safety, ii) automatic hierarchy learning (i.e., safety-first learning and then high-performance learning), and iii) safety-informed batch sampling to address the learning experience imbalance caused by corner cases. Experiments with a real quadruped robot, a quadruped robot in NVIDIA Isaac Gym, and a cart-pole system, along with comparisons and ablation studies, demonstrate the Real-DRL's effectiveness and unique features.

Summer is for relaxing, not cleaning. Upgrade your backyard setup with a robot that cleans your pool for you. As the owner of an above-ground pool, I can attest that cleaning leaves, bugs, and other detritus is an everyday chore if the cover is off. I put up with it since my pool is only in use seasonally, but WIRED reviewer Chris Null has an in-ground pool he uses year-round, so he does not have time for foolery like nets, brushes, and manual vacuums. He has been testing pool robots for years, and his favorite robot he's ever tried has been from Beatbot.

Learning robot control policies from human videos is a promising direction for scaling up robot learning. However, how to extract action knowledge (or action representations) from videos for policy learning remains a key challenge. Existing action representations such as video frames, pixelflow, and pointcloud flow have inherent limitations such as modeling complexity or loss of information. In this paper, we propose to use object-centric 3D motion field to represent actions for robot learning from human videos, and present a novel framework for extracting this representation from videos for zero-shot control. We introduce two novel components in its implementation.

We address key challenges in Dataset Aggregation (DAgger) for real-world contactrich manipulation: how to collect informative human correction data and how to effectively update policies with this new data. We introduce Compliant Residual DAgger (CR-DAgger), which contains two novel components: 1) a Compliant Intervention Interface that leverages compliance control, allowing humans to provide gentle, accurate delta action corrections without interrupting the ongoing robot policy execution; and 2) a Compliant Residual Policy formulation that learns from human corrections while incorporating force feedback and force control. Our system significantly enhances performance on precise contact-rich manipulation tasks using minimal correction data, improving base policy success rates by over 60% on two challenging tasks (book flipping and belt assembly) while outperforming both retraining-from-scratch and finetuning approaches. Through extensive real-world experiments, we provide practical guidance for implementing effective DAgger in real-world robot learning tasks.

NVIDIA introduces Halos for Robotics, which the company calls the industry's first full-stack safety system for robotics and physical AI operating near people.

Robotic task planning in real-world environments requires reasoning over implicit constraints from language and vision.

Howeverc,athesebinVLM-basedet above approaches remain limited in their support for diverse embodiment types.

A robot's instantaneous sensory observations do not always reveal task-relevant state information. Under such partial observability, optimal behavior typically involves explicitly acting to gain the missing information. Today's standard robot learning techniques struggle to produce such active perception behaviors. We propose a simple real-world robot learning recipe to efficiently train active perception policies.

Foundation models trained on web-scale data have revolutionized robotics, but their application to low-level control remains largely limited to behavioral cloning. Drawing inspiration from the success of the reinforcement learning stage in finetuning large language models, we propose a two-stage post-training approach for robotics.