Jogging and Delivery Robots

The OSU Dynamic Robotics Laboratory’s research team, led by Agility Robotics’ Jonathan Hurst, combined expertise from biomechanics and robot controls with new machine learning tools to accomplish something new: train a bipedal robot to run a full 5K on a single battery charge. This was an industry-first.

Cassie (2017-2019) was the previous robot of Agility Robotics. Cassie is a highly capable, rugged, bipedal research platform in use by numerous customers. With all-day battery life, an open architecture that provides access to low-level controls, and 3 years of included support, Cassie is the world’s best tool for pioneering new control methods for legged locomotion, and associated technologies. The final Cassie delivery occurred in July 2019 and it is no longer available for purchase.

Agility Robotics employees are applying this research to Digit which will help do last-mile deliveries and work in retail stores.

Agility Robotics has $28.8 million in funding and is partnering with Ford.

Digit is now on its third version. It addresses the mobility limitations of traditional robots, so that machines can work in environments designed for humans. Digit features robust walking and running gaits, perception to enable stair climbing and autonomous navigation in unstructured environments, and arms for basic manipulation tasks. Major features include:

Upper torso with integrated sensing, computing, and two 4-DOF arms

Extra bay for additional computing and custom payload integration

2-DOF feet for improved balance and stability on a wide variety of surfaces.

Sealed joints for all-weather outdoor operation.

UN 38.3 certified battery for cargo air shipment.

Comprehensive software API that leverages our controls, perception and autonomy algorithms to develop end-user applications

Low level API for customers wanting to develop their own controller

SOURCES- Agility Robotics
Written By Brian Wang,