NEW
ROBOTERA L7: The Fastest Humanoid Robot Redefining Agility and Dexterity
ArticlesRobotsCompaniesCountries
Contribute
Enquire about Robot

Simply fill up the form, our team will then reach out to you to connect to the relevant parties who can help facilitate your request.

Brand & Model
All fields required. Refer to our privacy policy for questions.
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
HomeRobots
Adam

Adam

Adam by PNDbotics is a 1.6m humanoid robot with 25 DOF, biomimetic design, and AI-driven reinforcement learning for agile, human-like locomotion and manipulation.
Enquire about robot
Specifications
Dev Doc
Website
Made by
PNDbotics (HK) Co., Limited
Software Type
Open Source
Software Package
Reinforcement learning and imitation learning algorithms. Real-time control and sensor fusion software. AI-powered motion simulation and training tools. Open-source SDK for customization and development.
Actuators
Adam uses 25 quasi-direct drive (QDD) force-controlled actuators distributed throughout its body, providing high torque (up to 340 N·m in legs), compliance, and energy efficiency for smooth, human-like motion.
Compiute
Powered by a 12th generation Intel i7 processor integrated with a real-time control unit (PND RCU), Adam manages actuator control, sensor data, battery management, and AI algorithms with low latency.
Sensors
Equipped with joint torque and position sensors, force sensors in legs, and optional vision modules. The sensor network is integrated via a high-speed Ethernet switch enabling precise feedback and control.
Max Op. time
mins
Available
China
Humanoid
2024
Pricing
Not provided.

Robot Brief

Adam, developed by PNDbotics, is a cutting-edge humanoid robot engineered for seamless human-robot collaboration across multiple industries. Standing 1.6 meters tall and weighing 60 kilograms, Adam features a highly biomimetic torso and pelvis design that enables unprecedented human-like agility and motion control. Utilizing 25 modular, quasi-direct drive (QDD) actuators with force control, Adam achieves remarkable dexterity and adaptability in dynamic environments. Its control system leverages reinforcement learning (RL) and imitation learning algorithms, allowing the robot to learn locomotion and complex movements through real-time interaction with its surroundings. Adam’s modular architecture simplifies maintenance and upgrades while supporting a wide range of applications including manufacturing, logistics, healthcare, and potentially space exploration. The integration of AI-powered motion simulation has significantly accelerated its development, reducing costs and enabling rapid iteration. PNDbotics’ open-source SDK and Nvidia Isaac Gym compatibility further empower developers to customize and extend Adam’s capabilities, setting a new standard in humanoid robotics.

Use Cases

Adam performs agile bipedal locomotion with human-like gait, capable of navigating complex terrains and maintaining balance with biomimetic pelvis mechanics. It executes dynamic and precise manipulation tasks, adapting to changing environments through reinforcement learning. Adam supports applications ranging from industrial automation and logistics to healthcare assistance, offering robust, energy-efficient operation and seamless human-robot interaction.

Industries

  • Manufacturing: Automates complex assembly and material handling tasks.
  • Logistics: Enhances warehouse operations with dexterous manipulation and navigation.
  • Healthcare: Assists in patient care and rehabilitation with adaptable motion.
  • Research & Development: Provides a versatile platform for robotics innovation and AI experimentation.
  • Space Exploration: Potential for deployment in challenging extraterrestrial environments.

‍

Specifications

Length
-
mm
Width
-
mm
Height (ResT)
-
mm
Height (Stand)
-
1600
mm
Height (Min)
mm
Height (Max)
mm
Weight (With Batt.)
-
60
kg
Weight (NO Batt.)
-
kg
Max Step Height
-
mm
Max Slope
+/-
-
°
Op. Temp (min)
-
°C
Op. Temp (Max)
-
°C
Ingress Rating
-
No items found.

Intro

Adam stands 1.6 meters tall and weighs 60 kg. It incorporates 25 degrees of freedom powered by modular quasi-direct drive actuators that provide high torque and force control. The robot’s biomimetic pelvis design enables natural and stable locomotion. Its torso mimics human anatomy, enhancing balance and agility. Adam’s control system is powered by an Intel i7 processor and a real-time network architecture (PDN) that integrates actuator control, battery management, and sensor data. The modular design allows for easy maintenance and upgrades. Optional dexterous hands and vision modules can be added for enhanced manipulation and perception. Adam supports AI-driven control algorithms, including reinforcement and imitation learning, enabling rapid adaptation and smooth motion across diverse tasks.

Connectivity

  • 16-port Gigabit Ethernet programmable switch for actuator and sensor network (PDN).
  • Wireless communication for remote monitoring and control.
  • Compatible with Nvidia Isaac Gym for simulation and AI training.
  • Open-source SDK for developer customization.

‍

Capabilities

  • Human-like bipedal locomotion with biomimetic pelvis mechanics.
  • 25 degrees of freedom for coordinated whole-body movement.
  • Force-controlled modular actuators for precise and compliant motion.
  • Reinforcement learning-based adaptive control for dynamic environments.
  • Energy-efficient operation with reduced mechanical wear.
  • Modular hardware for easy upgrades and customization.
  • Potential for integration with dexterous hands and vision systems.

‍