Deep Robotics Lynx M20 Pro

The Deep Robotics Lynx M20 Pro is a mid-sized, industrial-grade wheel-legged quadruped. It is described by Deep Robotics as the world's first wheeled-legged robot built specifically for challenging terrain and hazardous environments in industrial operation. By combining the rolling speed of wheels with the obstacle-crossing ability of legs, it transitions between surfaces without manual reconfiguration, covering ground efficiently while still handling stairs, slopes, debris, wetlands, and tunnels.

At 33 kg it is transportable by a single operator, and its compact form lets it work in spaces as narrow as 50 cm. It carries a 15 kg rated payload, with a maximum static load capacity of 50 kg, and runs up to three hours unloaded or 2.5 hours under load. Hot-swappable batteries support uninterrupted operation, and an optional self-charging capability is available.

Perception and autonomy are built on dual 96-line LiDAR units and dual wide-angle cameras, driving SLAM-based mapping, autonomous navigation, and omnidirectional obstacle avoidance. The platform is IP66-rated for dust and water, operates from -20 to 55 degrees Celsius, and offers an open architecture with Gigabit Ethernet, USB, and expansion power for adding sensors, payloads, or compute. It is built for power inspection, emergency response, logistics, and field research.

• Wheel-Leg Hybrid Locomotion: Rolls efficiently on flat ground and steps over obstacles when terrain demands, using Model Predictive Control and Reinforcement Learning to transition seamlessly between rolling and walking.
• Extreme Terrain Capability: Climbs continuous 25 cm stairs, clears single steps up to 80 cm, and handles slopes up to 45 degrees, with adaptive gait control and posture transitions for irregular ground.
• Industrial Endurance and Payload: Carries a 15 kg rated payload (50 kg maximum static load) and runs up to three hours unloaded, with hot-swappable batteries for uninterrupted missions.
• Autonomous Perception: Dual 96-line LiDAR units and dual wide-angle cameras provide a 360 by 90 degree field of view and roughly 860,000 points per second for SLAM, navigation, and omnidirectional obstacle avoidance.
• Built for Harsh Conditions: IP66-rated against dust and water, rated from -20 to 55 degrees Celsius, with bidirectional lighting for operation in darkness.
• Open and Expandable: Gigabit Ethernet, USB 3.0, OTA updates, and 24/72 V expansion power up to 300 W support third-party sensors, payloads, and compute modules.

The Lynx M20 Pro is built for industrial and field operations where terrain access, autonomy, and ruggedness are critical. In power inspection, it navigates substations, transmission corridors, and electrical infrastructure autonomously, reducing human exposure in high-voltage environments while collecting consistent inspection data. In emergency response, its wheel-leg design reaches rubble, steep slopes, and narrow access points that conventional wheeled platforms cannot.

It also serves logistics tasks across uneven sites and supports research and development in quadruped locomotion, SLAM, autonomous navigation, embedded AI, and sensor fusion. Its open architecture and mounting interfaces let integrators add LiDAR, cameras, a robotic arm, or application-specific payloads. This is a professional platform intended for trained operators and integrators rather than consumer use.

1. Unbox and inspect: Remove the Lynx M20 Pro and accessories, and inspect for any shipping damage before powering on.
2. Install and charge the battery: Insert a charged hot-swappable battery, or charge a battery fully (about 1.5 hours) before first use.
3. Power on and connect: Power the unit and connect over Gigabit Ethernet or the supported wireless link, following the user manual.
4. Update software: Apply any available OTA update before field operation.
5. Calibrate and test: Run the startup checks, confirm LiDAR and camera perception, and test movement in a safe, open area before deploying on terrain.
6. Review safety guidance: Confirm operators are trained and follow Deep Robotics' operating limits for speed, slope, and load.