World's First Multi-Modal Biped Robot. Point-Foot, Sole & Wheeled. Open SDK. Python-Native. The LimX Dynamics TRON 1 EDU is the world's first multi-modal bipedal robot, engineered specifically to advance humanoid reinforcement learning research. Its patented "Three-In-One" modular foot-end design enables seamless transitions between Point-Foot, Sole, and Wheeled locomotion configurations — allowing researchers to develop, test, and validate multi-modal locomotion algorithms across all three modes on a single hardware platform. With a fully open SDK, Python-native development support, and compatibility with NVIDIA Isaac, MuJoCo, and Gazebo simulation environments, TRON 1 EDU is the most accessible and versatile bipedal research platform available. Contact the Maverick team for pricing and availability. Three-In-One Modular Foot-End System Point-Foot — Simplified legged form factor for fundamental locomotion studies and basic control algorithm development Sole — Humanoid legged form enabling standing and walking research for human-like gait pattern development Wheeled — All-terrain wheeled mobility for enhanced speed, efficiency, and surface traversal capability Automatic hardware recognition and software adaptation enable fast, seamless switching between configurations with minimal downtime between experiments. Key Features Ready-to-Use Deployment — Built-in high-performance motion control algorithms enable immediate deployment for research and development without extensive setup or calibration. Fully Open SDK & Hardware Interface — Complete SDK and hardware interface access supports high-complexity algorithm validation and custom integration for both novice and experienced robotics developers. Python-Native Development — Full-process development in Python eliminates the need for C++, lowering the barrier to entry for researchers and enabling faster iteration cycles. Simulation Platform Compatibility — Compatible with NVIDIA Isaac, MuJoCo, and Gazebo — minimizing the Sim2Real gap for algorithms developed in simulation before hardware deployment. Quick Assembly & Automatic Adaptation — Efficient foot-end switching with automatic hardware recognition and software adaptation ensures seamless transitions between locomotion modes. Designed For Humanoid reinforcement learning research programs Multi-modal locomotion algorithm development and validation University and institutional robotics research AI and robotics education programs Sim2Real transfer research with NVIDIA Isaac, MuJoCo, and Gazebo
