Summary

  • Robotics firm Boston Dynamics is looking to accelerate its robot’s learning ability through the application of machine learning.
  • Founder and chairman Marc Raibert said the company’s parkour and dancing robots are learning to replicate behaviours without the need for human interaction.
  • advancement in machine learning has sped up the learning process in the robots through accurate simulations that allows them to practice in a virtual environment.
  • Boston Dynamics used a technique called reinforcement learning to improve its four-legged robot called Spot, as well as its human-like robot, Atlas.
  • The robots are being seen as a potential answer to difficulties faced by wheels in certain terrain, such as oil rigs and building sites.

By Will Knight

Original Article