What's in This Configuration

The R1-A7-D Smart H represents the biomimetic dexterity platform for research-grade manipulation. It pairs the R1-A7-D autonomous mobile base — complete with 7-DOF arms, MID360 LiDAR, stereo vision, and a removable chassis battery — with tendon-driven dexterous hands on both arms. These hands replicate the human hand's tendon-actuated architecture, delivering fluid, compliant motion and natural grasping behaviour through cable-driven transmission. With no wrist cameras or tactile sensors, the Smart H focuses purely on kinematic dexterity and biomimetic control, making it ideal for research in tendon-driven mechanisms, compliant manipulation, and human-like hand motion.

Tendon-Driven Dexterous Hands — Biomimetic Cable-Actuated Dexterity

The tendon-driven dexterous hands on the Smart H replicate the human hand's biological actuation architecture: motors in the forearm pull cables (tendons) that route through the wrist and into each finger joint, mimicking the way human muscles and tendons control the hand. This cable-driven design delivers inherently compliant motion — the fingers yield gracefully to external forces, making the hand safe for contact-rich interaction and capable of adaptive grasping without force sensors. The distributed tendon routing enables independent control of each finger joint, supporting the full taxonomy of human grasps: power grips, precision pinches, tripod grasps, lateral holds, and in-hand manipulation. Unlike rigid gear-driven hands, tendon-driven mechanisms exhibit back-drivability and passive compliance, essential for research in safe human-robot interaction, variable stiffness control, and biomimetic manipulation.

Biomimetic Grasping — Human-Like Compliance & Dexterity

Smart H unites 7-DOF arm redundancy with tendon-driven hand compliance for human-like manipulation. The cable-actuated fingers can modulate stiffness by co-contracting antagonistic tendon pairs — just like human muscles — enabling the robot to switch seamlessly between rigid precision holds and soft, enveloping power grasps. This variable stiffness behaviour is impossible to achieve with conventional gear-driven actuation and opens research frontiers in impedance control, grasp force modulation without force sensors, and safe physical interaction. The tendon routing also allows the fingers to passively adapt to object geometry: when contacting an irregular surface, the compliant joints deflect naturally, distributing contact forces evenly across the hand. Combined with the 555 mm arm reach and 120 Nm shoulder torque, the Smart H can execute whole-arm manipulation strategies that leverage both arm redundancy and hand compliance.

Stereo Vision Perception — Pure Visual Manipulation

Smart H relies on pure stereo vision for manipulation guidance. The head-mounted binocular camera provides global scene understanding, object recognition, and depth estimation, while the 7-DOF arm redundancy enables the robot to position the hands optimally within the visual field. Without wrist cameras or tactile sensors, the Smart H represents a minimal-sensing, maximum-dexterity configuration — ideal for research that isolates kinematic and dynamic control from multi-modal perception. The Orin Nano 40 TOPS processes stereo depth, visual servoing, and arm trajectory planning entirely onboard, demonstrating that advanced manipulation is achievable with vision alone when paired with sufficiently dexterous, compliant end effectors.

7-DOF Arms — Human-Like Redundancy & Reach

The R1-A7-D features seven degrees of freedom per arm — matching the kinematic complexity of the human upper limb. With a 555 mm reach (forearm + upper arm) and 120 Nm peak shoulder torque, these arms can reach around obstacles, maintain orientation while translating the end effector, and execute complex trajectories impossible for simpler 5- or 6-DOF manipulators. Crossed-roller and double-row ball bearings on every joint ensure smooth, precise motion with minimal backlash. Dual encoders per joint provide closed-loop accuracy for research-grade repeatability.

Autonomous Mobile Base — Navigate, Then Manipulate

The R1-A7-D chassis adds 3 degrees of freedom to the upper body: a telescopic pillar for height adjustment and a 2-DOF wheeled base for omnidirectional navigation. Combined with the MID360 LiDAR and stereo camera head, the robot builds real-time maps, plans paths, and avoids obstacles autonomously — then stops precisely to manipulate. The removable lithium-ion battery in the chassis delivers approximately 1.5 hours of untethered operation. Please note: the upper body of this model does not accept a battery.

LiDAR + Stereo Vision — Complete Perception Stack

Smart H processes data from two complementary sensors: a MID360 360° LiDAR for metrically accurate mapping and long-range obstacle detection, and a binocular camera head (146°×124° FOV, 60 mm baseline) for colour vision, object recognition, and depth estimation. Fused on the Orin Nano 40 TOPS, this dual-modality perception stack supports research in visual-LiDAR SLAM, semantic navigation, and active perception for mobile manipulation.

Orin Nano 40 TOPS — Onboard AI for Navigation & Control

The built-in NVIDIA Jetson Orin Nano delivers 40 TOPS of AI compute, running SLAM, path planning, object detection, and arm trajectory generation locally. No external GPU server, no cloud dependency — just power on and deploy. The module is pre-installed and thermally integrated with the chassis cooling system. Optional 40 or 100 TOPS compute expansion modules are available for future upgrades.

Binocular Vision & Multi-Camera Perception

A binocular camera head delivers a 146° × 124° field of view with a 60 mm stereo baseline for scene-level understanding. RGB streams at 1280×720@30 Hz and depth at 544×448@10 Hz. Combined with the MID360 LiDAR, the R1-A7-D Smart H offers a complete perception stack for research in visual servoing, semantic navigation, and vision-guided compliant manipulation with tendon-driven hands.

Voice Collaboration & Natural Interaction

Four-array beamforming microphones and dual 3 W speakers enable clear voice capture and synthesis in noisy labs, warehouses, or classrooms. The open audio framework supports custom wake words, multilingual TTS, and integration with LLMs for conversational command-and-control — even while the base is in motion.

Collapsible. Mobile. Ready to Deploy.

At 683 × 520 × 440 mm collapsed (1323 mm elevated), the R1-A7-D Smart H navigates standard doorways, elevators, and lab aisles. The telescopic pillar lowers the upper body for transport and raises it for manipulation. At approximately 32 kg with battery, it is a self-contained mobile research platform — no external compute rack, no tether cables during operation.

Full-Stack Open Development

Unitree provides a mature, Linux-based robot development framework with open APIs for the underlying system, robotic arms, mobile base, audio, lighting, and visual control. ROS 2 native support, dual encoders on every arm joint, hollow internal wiring, and low-inertia, high-speed inner-rotor permanent magnet synchronous motors make the R1-A7-D Smart H a clean slate for reinforcement learning, tendon-driven hand control research, compliant manipulation, embodied AI, and autonomous navigation studies. The built-in Orin Nano 40 TOPS handles perception and planning locally without external hardware.

Upgrade Path — Grow With Your Research

Smart H is the biomimetic dexterity configuration. Expand capability as your research evolves:

• End Effectors — Swap to LinkerBot O6 for high-DOF gear-driven dexterity, BrainCo Revo 2 for anthropomorphic grasping, or Dex3-1 for three-finger adaptive manipulation.
• Wrist Cameras — Add wrist-mounted RGB cameras for close-up visual feedback during tendon-driven grasping and visual servoing.
• Tactile Sensing — Integrate force-sensitive tactile arrays for contact-aware manipulation with compliant tendon-driven fingers.
• Large VLM — Deploy vision-language models for natural-language task specification and scene understanding with biomimetic hands.
• Compute — Swap to the 100 TOPS expansion for large-model inference, multi-modal fusion, tendon dynamics simulation, and multi-robot coordination.

Designed For

Technical Specifications

Model	Unitree R1-A7-D Smart H (Mobile Base)
Dimensions (Collapsed)	683 × 520 × 440 mm (H × W × D)
Dimensions (Elevated)	1323 × 520 × 440 mm (H × W × D)
Weight (with Battery)	~32 kg
Total DOF	20 (Arm 7×2, Waist 1, Head 2, Pillar 1, Base 2)
Arm Reach	555 mm (forearm + upper arm)
Peak Shoulder Torque	120 Nm
Max Arm Payload	~2 kg (posture-dependent)
End-Effector Precision	±0.1 mm
End Effectors	Tendon-driven dexterous hands (both arms, pre-installed)
Hand Actuation	Cable-driven tendon transmission with antagonistic routing
Hand Compliance	Passive compliance via elastic tendon coupling; variable stiffness via co-contraction
Wrist Cameras	Not included (optional)
Tactile Sensing	Not included (optional)
Vision	Binocular camera — RGB 1280×720@30 Hz, Depth 544×448@10 Hz, FOV 146°×124°
LiDAR	MID360 360° LiDAR (chassis-integrated)
Head Compute	8-core high-performance CPU + 10 TOPS NPU
Body Compute	8-core high-performance CPU
Onboard AI Compute	NVIDIA Jetson Orin Nano 40 TOPS (built-in)
Audio	4-array microphone + dual 3 W speakers
Connectivity	Wi-Fi 6, Bluetooth 5.2, OTA updates
Power	Removable chassis Li-ion battery (~1.5 h) + external DC PSU. Upper body does not accept a battery.
Joint Bearings	Crossed-roller + Double-row ball bearings
Joint Motor	Low-inertia, high-speed, inner-rotor permanent magnet synchronous motor
Joint Encoders	Dual + Single per joint
Cabling	Hollow + Internal routing
Cooling