Thoth develops AI-powered robotic solutions for advanced process automation in smart factories. Their technology enables flexible automation capable of handling uncertainty across diverse manufacturing tasks, from inspection and assembly to complex material handling. The core offering utilizes deep learning and autonomous programming to deploy adaptable robotic systems for high-mix, high-volume production environments.
Funding
Funding not disclosed
Founders
Product
Problem
Many industries require flexible and precise automation solutions to handle complex tasks such as EV battery disassembly, multi-product packaging, and quality control. Traditional automation systems often lack the adaptability to manage diverse product types and rapidly changing production requirements. This results in inefficiencies and increased operational costs.
Solution
THOTH provides AI-powered robotic solutions designed for process automation across various industries. Their technology leverages deep learning and reinforcement learning to enable robots to perform complex tasks with minimal human intervention. The Robot AI-based Autonomous Programming Solution (RAAPS) allows robots to adapt to new tasks through operator demonstration, physics simulation-based learning, and transfer learning. This approach enables automation of processes previously considered too difficult or costly to automate, such as electric vehicle battery recycling, logistics packaging, and custom manufacturing processes.
Target Audience
THOTH's primary customers are manufacturers and logistics providers seeking to automate complex processes, improve efficiency, and reduce operational costs in industries such as recycling, electronics, and general manufacturing.
Features
- AI-powered robotic arms for precise and adaptable automation
- RAAPS (Robot AI-based Autonomous Programming Solution) for autonomous task adaptation
- Operator Demonstration (Imitation Learning & Inverse Reinforcement Learning) for intuitive robot teaching
- Physics Simulation-based Learning (Reinforcement Learning & Self-Learning) for robust performance in dynamic environments
- Transfer Learning & Real-time Inference and Control for rapid deployment across different tasks
- Solutions for EV battery module, cell, and pack disassembly
- Capabilities for multi-product small/large batch packaging and palletizing/depalletizing
- Visual inspection and classification of products and parts