Proception develops advanced humanoid hands that integrate robotics and artificial intelligence to achieve human-like dexterity in machines. Their core product, ProHand, aims to redefine the capabilities of robotic manipulation across various applications. This technology offers enhanced precision and versatility for complex tasks currently limited by conventional robotic end-effectors.
Funding
Funding not disclosed
Founders
Product
Problem
Existing robotic hands lack the dexterity, precision, and adaptability required for complex manipulation tasks in research and industrial settings. Traditional robotic systems struggle to replicate the fluidity and finesse of human hand movements, limiting their application in tasks requiring fine motor skills and delicate object handling.
Solution
Proception AI develops advanced humanoid robotic hands, called ProHand, designed to replicate human-like dexterity and precision. ProHand utilizes advanced mechanical design, AI-powered control systems, and integrated sensors to achieve precise manipulation, adaptive grip, and dynamic movement. The robotic hand features tactile sensing, force control, and joint flexibility, enabling it to handle fragile objects and adapt to different object shapes. Proception's approach to data collection, gathering data through human interaction rather than teleoperation, enhances the efficiency of AI model training for manipulation and task execution. This allows ProHand to perform complex tasks with a high degree of accuracy and adaptability.
Target Audience
The primary target audience includes research institutions and industrial clients requiring advanced robotic manipulation capabilities for applications in healthcare, manufacturing, and other sectors.
Features
- 20+ degrees of freedom with tendon-driven actuation for human-like finger movement
- High-resolution touch sensors for precise object interaction and tactile feedback
- AI-powered control systems for real-time adaptation and learning from experience
- Adaptive grip strength that automatically adjusts to different object shapes and sizes
- Force control for handling fragile objects with delicate precision
- Real-time learning capabilities to improve performance over time
- Multi-mode operation for versatile task execution