Covariant Unveils AI-Powered Warehouse Robots

Pieter Abbeel’s new robotics startup Covariant this week deployed their AI-equipped robot at customer facilities in North America and Europe in the apparel…

NVIDIA logo
NVIDIA
•
Nefi Alarcon
•2 min read•advanced•
--
•View Original
PyTorch

Overview

Covariant has launched AI-powered warehouse robots that are currently deployed in customer facilities across North America and Europe, specifically in the apparel, pharmaceutical, and electronics industries. The robots utilize deep learning models and GPU acceleration to enhance their operational capabilities, learning from one another to improve performance continuously.

What You'll Learn

1

How to leverage AI-powered robots for warehouse automation

2

Why reinforcement learning is crucial for robotic task completion

3

How to implement deep learning models using NVIDIA GPUs

Prerequisites & Requirements

  • Understanding of AI and machine learning concepts
  • Familiarity with NVIDIA GPUs and deep learning frameworks like PyTorch(optional)

Key Questions Answered

How do Covariant's robots learn to manipulate objects?
Covariant's robots utilize reinforcement learning models, which allow them to train through trial and error until they successfully learn to manipulate objects. This approach enables them to develop general abilities such as robust 3D perception and real-time motion planning.
What technologies are used in Covariant's robots?
Covariant's robots are equipped with NVIDIA GeForce 2080Ti and Quadro RTX 6000 GPUs for training and inference. They also utilize the cuDNN-accelerated deep learning PyTorch framework for their backend processing.
What industries are Covariant's robots deployed in?
Covariant's AI-powered robots are currently deployed in the apparel, pharmaceutical, and electronics industries across customer facilities in North America and Europe.
What is the accuracy of Covariant's AI-based system in item detection?
Covariant's AI-based system can detect over 10,000 different items with an accuracy of 99%, demonstrating its effectiveness in warehouse operations.

Key Statistics & Figures

Item detection accuracy
99%
Covariant's AI-based system can detect over 10,000 different items with this level of accuracy.

Technologies & Tools

Some links below are affiliate links. We may earn a commission if you make a purchase.

Hardware
Nvidia Geforce 2080ti
Used for training deep learning models.
Hardware
Nvidia Quadro Rtx 6000
Used for inference, processing sensory inputs, and performing object detection.
Software
Pytorch
The deep learning framework used for developing the AI models.

Key Actionable Insights

1
Integrating AI-powered robots into warehouse operations can significantly enhance efficiency and accuracy.
By automating tasks such as item picking and order fulfillment, businesses can reduce labor costs and improve order accuracy, leading to better customer satisfaction.
2
Utilizing reinforcement learning in robotics allows for continuous improvement and adaptation to new tasks.
This approach enables robots to learn from their experiences, making them more versatile and capable of handling a wider range of operations without explicit programming.
3
Employing NVIDIA GPUs for deep learning tasks can accelerate model training and inference processes.
The use of high-performance GPUs allows for faster processing of sensory inputs and object detection, which is crucial for real-time robotic applications.

Related Concepts

Reinforcement Learning
Deep Learning
Robotics In Warehouse Automation