.jpg%3Ftable%3Dblock%26id%3Db08f75f4-4e02-4fdf-8421-af20663b7851%26cache%3Dv2&w=3840&q=75)
Hi, We are LiOps
LiOps develops AI Robotics Foundation Model called ”RT3D” (Robotics-Transformer-3D-Diffusion) using the latest 3D deep learning technology.
We dream of a world where humanity break free from repetitive physical labor through our solutions, enabling more people to engage in more productive work.
LiOps Story
"Why is it still difficult to commercialize AI robotics solutions?"
LiOps is a deep tech startup that started with this question.
One of the main reasons for the difficulty in commercializing AI robotics is the limitation of data. Currently, most AI models are trained on data optimized for specific (niche) tasks, making it challenging to apply them universally. This conflicts with the real-world requirements where robots need to respond flexibly to various environments and tasks.
LiOps aims to solve this problem in an innovative way. Our approach is as follows:
✅ Universal data collection: We collect extensive and diverse robot task and environment data that is not limited to specific tasks.
✅ Integrated learning method: All collected data is comprehensively learned in one large foundation model.
✅ Strengthening generalization ability: Through this approach, the model develops the ability to understand not only specific tasks but also overall robot movements and environments.
✅ Flexible application: As a result, our foundation model can perform various niche tasks with a single model.
This approach is fundamentally different from existing narrow-scope AI models. We aim to develop true 'intelligence' that allows robots to adapt to various situations and easily learn new tasks, just like humans.
LiOps' foundation model will bring a paradigm shift to the robotics industry. This goes beyond simply improving efficiency, making robots more useful and adaptable companions in our daily lives and industrial settings."
3D vision-based AI robotics is experiencing explosive growth
3D vision-based AI robotics is experiencing explosive growth, driven by rapid advancements in technology and increasing demand across various industries. This growth is characterized by several key factors:
- Technological convergence: The combination of 3D vision systems, artificial intelligence, and robotics is creating powerful new capabilities for automated systems.
- Market expansion: The global robotic vision market is projected to grow from $2.6 billion in 2023 to $4.0 billion by 2028, indicating strong industry momentum.
- Improved accuracy and efficiency: AI-powered 3D vision systems are enhancing robots' ability to perceive and interact with their environment, leading to unprecedented levels of speed, accuracy, and payload handling capabilities.
- Widening applications: These technologies are finding new uses beyond traditional manufacturing, entering sectors such as healthcare, logistics, retail, and construction.
Companies and research institutions worldwide are at the forefront of this revolution.
In the United States, Covariant.ai, founded by UC Berkeley researchers, is pushing the boundaries of AI in robotics. Their development of a Robotic Foundation Model (RFM) aims to create a versatile AI capable of adapting to various tasks and environments, particularly in logistics and warehouse automation.
In South Korea, CMES is making significant strides by integrating AI, 3D vision, and robot guidance technologies to create intelligent robotic automation solutions. Their focus on enhancing production efficiency across major industrial sectors showcases the practical applications of this technology.
These companies are just two examples in a global landscape of innovation. From established tech giants to innovative startups and academic institutions, there's a worldwide effort to advance 3D vision-based AI robotics. This collective push is driving the field forward, promising transformative changes in how robots perceive, understand, and interact with the world around them.As this technology continues to evolve, we can expect to see even more sophisticated and capable robotic systems emerging, further revolutionizing industries and opening up new possibilities for human-robot collaboration.
3D Gaussian Splatting
“Why now, and Why LiOps?”
Now is the perfect time for LiOps to revolutionize the robotics industry with our innovative foundation model approach. Here's why:
- Convergence of expertise: Our co-founders bring unique and complementary expertise crucial for this breakthrough. Co-founder Seon's experience with Hyundai Mobis in pioneering deep learning for autonomous driving, coupled with his world-leading expertise in LiDAR-based 3D segmentation, provides the backbone for our advanced 3D perception capabilities. Co-founder Park's background in NeRF-based 3D reality reconstruction and successful MLOps implementation brings the critical ability to create and manipulate highly accurate virtual environments.
- Overcoming data limitations: Unlike traditional AI robotics solutions that struggle with limited, task-specific datasets, LiOps is uniquely positioned to overcome this challenge. We're creating a vast, diverse dataset that combines real-world data from various robotic applications with synthetic data generated through our advanced simulation capabilities. This approach allows us to train our foundation model on a scale previously unattainable in robotics.
- Unified foundation model: LiOps is pioneering the development of a single, versatile foundation model for robotics. This model can understand and generate 3D spatial information directly, eliminating the need for intermediary 2D image processing. Our approach enables the model to generalize across a wide range of tasks and environments, something that current niche-focused AI solutions cannot achieve.
- Bridging simulation and reality: By leveraging our expertise in both 3D perception and reality reconstruction, we can create highly accurate simulations that closely mirror real-world conditions. This capability allows for extensive training and testing in virtual environments, significantly accelerating development and reducing costs associated with real-world robot deployment.
- Addressing industry-wide challenges: Our foundation model approach directly addresses key limitations in current AI robotics solutions, including the lack of generalization, high implementation costs, and integration difficulties. By providing a single, adaptable model, we can significantly reduce the complexity and cost of deploying robotic solutions across various industries.
- Timing and market readiness: The robotics industry is at a tipping point, with increasing demand for flexible, intelligent solutions across multiple sectors. Our approach aligns perfectly with this need, offering a scalable, adaptable solution that can be applied to a wide range of robotic applications.
Only LiOps has the unique combination of expertise, vision, and technological approach to develop this groundbreaking robotics foundation model. Our solution promises to replace countless manual tasks with a single, highly adaptable AI system, ushering in a new era of robotic capabilities and applications. By leveraging our founders' deep expertise in 3D perception, reality reconstruction, and AI, we are uniquely positioned to create a truly transformative solution for the robotics industry.
AMR Solution
