As embodied intelligence becomes one of the most widely discussed topics at this year's WAIC, the robotics industry is entering a new stage of development.
The recently concluded 2026 World Artificial Intelligence Conference (WAIC) brought together cutting-edge technologies and innovative achievements from the global AI field. From foundation models to intelligent terminals, from robots to industrial applications, "embodied intelligence" has undoubtedly become one of the hottest keywords at the event.
SLAMTEC Founder and CEO Shikai Chen was invited to attend the "Embodied Humanoid Robot Deployment and Core Component Development Forum", where he delivered a keynote speech and participated in a roundtable discussion, joining industry partners to explore the future development directions of core robotics technologies in the era of embodied intelligence.
Focusing on Embodied Intelligence: Fundamental Robot Capabilities Are Entering a New Stage of Evolution
In his keynote speech titled "Building Multimodal Spatial Perception and Autonomous Action Platforms for the Era of Embodied Intelligence", Shikai Chen shared SLAMTEC's perspectives on building fundamental capabilities for embodied intelligence, focusing on the development trends of robot spatial perception and mobile platforms.
Shikai Chen stated:
Currently, wheeled and legged robots have become the two major technological routes for embodied intelligent robots. Among them, wheeled robots, which combine high mobility efficiency and flexible application capabilities, remain an important direction for industrial deployment at the current stage. However, traditional wheeled robots still face many challenges.
-
Limited degrees of freedom, making it difficult to adapt to narrow and complex environments
-
Reliance on 2D laser navigation, making it difficult to meet the requirements of 3D environmental perception
-
Traditional SLAM algorithms are increasingly unable to support the requirements of environmental understanding and intelligent decision-making in the era of AI foundation models
Building on years of accumulated expertise in core technologies, SLAMTEC has developed fu
ll-stack capabilities integrating LiDAR, multimodal perception algorithms, and intelligent mobile robot platforms. To address the advanced deployment requirements of humanoid robots, SLAMTEC's newly launched
Poseidon omnidirectional mobile robot platform has become a key technical benchmark showcased during the speech.
Omnidirectional Mobility, Breaking Through the Limitations of Traditional Movement
Adopting an industry-leading four-wheel four-steering / four-wheel eight-drive architecture, it supports lateral movement, in-place rotation, diagonal driving, and parking modes, completely solving the problems of rigid movement and complicated steering found in traditional mobile platforms, while meeting the flexible mobility requirements of narrow spaces and complex operating environments.
Multi-Sensor Fusion, Enabling Human-Robot Collaboration Scenarios
Equipped with front and rear dual LiDAR sensors as standard (optional 3D multi-line LiDAR), dual depth vision cameras, and ultrasonic sensors, it builds a 360° omnidirectional perception system. Combined with SLAMTEC's self-developed SLAM algorithms, it enables full-scene 3D perception and precise obstacle avoidance, supporting advanced human-robot collaboration and autonomous operations in complex environments.
SLAMWARE NX: A New Generation of Self-Learning Localization Engine
Powered by the dedicated embodied intelligence version of the upgraded Slamware NX navigation system, it achieves a comprehensive evolution from 2D LiDAR SLAM to vision-LiDAR fusion 3D SLAM: through online autonomous learning and correction, it continues to provide stable and reliable localization even when environmental changes exceed 80%.
Flexible Adaptation Across All Forms, Accelerating Development and Deployment
With a fully modular architecture, stable payload capability, and flexible integration options, it supports multiple configurations including bare mobile platforms, customized enclosures, and complete robot systems. It is compatible with various upper structures such as folding mechanisms, lifting systems, and lead-screw structures. Available in both standard and ultra-low center-of-gravity versions, with a minimum installation height of only 65mm, it meets diverse customized development requirements.
Based on years of accumulated SLAM navigation expertise, the platform integrates full-stack capabilities including mapping, path planning, intelligent obstacle avoidance, and multi-robot scheduling. Native ROS/ROS2 compatibility and a mature SDK significantly reduce the overall robot development cycle to just one month, greatly lowering the barrier to development.
Beyond mobility capabilities, a robot's ability to "understand its environment" is equally fundamental to embodied intelligence.
The Aurora series integrates binocular vision, deep learning algorithms, and SLAM technology to enable high-precision localization, real-time 3D mapping, and semantic perception. It helps robots evolve from "seeing space" to "understanding space", further enhancing their autonomous perception and decision-making capabilities in complex dynamic environments.
Shikai Chen stated:
The future development of embodied intelligence will not be a competition between individual products, but rather a systematic integration of fundamental capabilities such as spatial perception, localization and navigation, and mobile platforms.
High-Level Dialogue: Exploring the Path Toward Independent and Controllable Development of Domestic Robotics
In addition to the keynote speech, Shikai Chen also participated in the forum's roundtable discussion, engaging in in-depth exchanges with guests from across the robotics industry supply chain on the topic of "How Can Domestic Humanoid Robots Achieve Independent Control of Core Components and Scale Deployment?"
In response to the moderator's question regarding "the large-scale deployment and global expansion of domestic humanoid robots, with a focus on supply chain autonomy, bottlenecks in core components, and domestic substitution", Shikai Chen from SLAMTEC shared his insights:
Regarding LiDAR: In the early stages, the LiDAR industry was completely dominated by overseas companies, which was one of the core reasons behind SLAMTEC's founding and commitment to independent R&D and mass production. Today, we have achieved major breakthroughs. The sensor industry has completely overcome the challenge of technological dependency, with mature domestic alternatives and broad industry applications.
Regarding AI and embodied intelligence, the current industry landscape continues to improve. Among SLAMTEC's current products, whether it is the universal mobile platform developed for the embodied intelligence robotics industry or the 3D localization and mapping solutions designed for complex environments and embodied intelligence applications, the localization rate of domestic components has exceeded 99%.
At the same time, Shikai Chen also pointed out the key factor that will truly constrain the long-term development of the industry:
The lack of influence in establishing industry standards remains a core challenge that we need to address over the long term.
This is also the core reason why SLAMTEC continues to actively engage in the development of national and industry association standards. Over the years, SLAMTEC has actively fulfilled its industry responsibilities by deeply participating in the formulation of multiple standards covering intelligent robot core components, algorithm evaluation, and application scenarios. Through filling gaps in industry standards and establishing unified technical and evaluation frameworks, SLAMTEC aims to drive standardized and high-quality development across the robotics industry.
When discussing "When will China's humanoid robots, driven by cost-performance advantages, achieve large-scale market adoption? And which types of humanoid robots, in what scenarios, and for what tasks will achieve market breakthroughs first?", Shikai Chen expressed a more optimistic view of the industrialization prospects of domestic embodied intelligent robots compared with the more conservative outlook commonly seen in the industry. He believes:
China has unique advantages in technology deployment: industrial application scenarios show greater acceptance of emerging technologies and a stronger willingness to experiment, while the speed of technology implementation and iteration significantly exceeds that of overseas markets.
The industry's inflection point is approaching. Previous breakthroughs in robotic vacuum cleaners and delivery robots were driven by the maturity of underlying technologies. Today, with the rapid iteration of domestic embodied intelligence technologies and locally developed foundation models, along with continuous improvements in both hardware and software ecosystems, the industry has already established the foundation for large-scale deployment.
From an industry development perspective: the widespread adoption of embodied intelligent robots among the general public may still require 3-5 years, but large-scale deployment in specific application scenarios is expected to begin in the second half of next year. From the second half of this year through next year, perception hardware, mobility solutions, and foundation model-based solutions will gradually mature and cover the needs of most application scenarios. Combined with future computing power upgrades and a significant reduction in hardware costs in 2027, the barriers to industrial deployment will continue to decrease.
Regarding application scenarios: flexible manufacturing and long-tail scenarios in small and medium-sized factories will achieve large-scale deployment first, outperforming traditional high-precision industrial scenarios. These scenarios feature higher fault tolerance, emphasize flexibility rather than extreme precision, and are highly compatible with the advantages of embodied intelligence robots in terms of high adaptability and low cost, making them key areas for the first wave of large-scale humanoid robot deployment.
SLAMTEC Remains Committed to Building the "Fundamental Capabilities" of Robotics
In recent years, artificial intelligence has been accelerating its transition from the digital world into the physical world, with robots becoming a key vehicle for this transformation.
The more complex the application scenarios become, the greater the need for stable and reliable localization and navigation capabilities, more precise spatial perception capabilities, more open and easy-to-use intelligent robot mobile platforms, and solutions that can be deployed faster.
In the future, SLAMTEC will continue working together with industry partners to explore innovations in core robotics technologies, jointly driving embodied intelligence into more real-world application scenarios and enabling robots to truly become an important bridge connecting artificial intelligence with the physical world.
Although WAIC 2026 has come to an end, the era of embodied intelligence is rapidly approaching. Driven by technological innovation, SLAMTEC will continue moving forward together with industry partners toward the next stage of the robotics industry.
WAIC 2026 | Focusing on the Fundamental Capabilities of Embodied Intelligence, SLAMTEC Builds Competitiveness for Domestic Embodied Intelligence Through Spatial Perception and Intelligent Mobility Technologies
WAIC 2026 | Focusing on the Fundamental Capabilities of Embodied Intelligence, SLAMTEC Builds Competitiveness for Domestic Embodied Intelligence Through Spatial Perception and Intelligent Mobility Technologies
