Skip to main content
H

Hillbot

Hillbot develops embodied AI technology to advance general-purpose robotics through skill acquisition. The company leverages real-world data and synthetic data generated via simulation and 3D generative AI for robust robotic learning. Their platform enables the training of foundational models for whole-body control across various robotic applications.

La Jolla, United StatesFounded 2024271K+ followers
Updated 3 months ago

Funding

Funding not disclosed

Funding rounds are not available yet.

Founders

Product

Problem

Training robots to perform tasks in unstructured, real-world environments is challenging due to the difficulty of acquiring sufficient real-world data and the limitations of traditional robot learning techniques. The sim-to-real gap further complicates the process, as skills learned in simulation often fail to transfer effectively to physical robots.

Solution

Hillbot provides embodied AI solutions that enable robots to navigate and operate in complex, unstructured environments. The company leverages generative AI and simulation to create vast amounts of synthetic training data, which is then used to train foundational models for robotics. This approach allows robots to learn a wide range of skills and adapt to new situations more effectively. Hillbot's core technology, MANISKILL, is an open-source platform for robot simulation that allows organizations to harness the power of infinite synthetic training data. SAPIEN, another key technology, offers a unified framework for robot simulation and scalable synthetic training data generation.

Target Audience

Hillbot's primary customers are organizations involved in robotics research and development, including those focused on industrial and domestic robot applications.

Features

  • MANISKILL: Open-source platform for robot simulation, enabling the generation of infinite synthetic training data.
  • SAPIEN: Unified framework for robot simulation and scalable synthetic training data generation.
  • 3D Generative AI: Generates diverse arrangements of objects across scenes and environments to ensure synthetic training data diversity.
  • Robot Learning: State-of-the-art robot learning for whole-body control of robots.
  • NVIDIA Cosmos integration: Utilizes the NVIDIA Cosmos platform to accelerate the development of physical AI systems.
This profile is AI-generated and may contain inaccuracies.