Proving the Business Case for the Internet of Things

Ford and Agility develop robot for making deliveries

Steve Rogerson
June 4, 2019

Ford is working with Agility Robotics to develop a robot that changes the way self-driving cars make deliveries. Digit is a two-legged robot that sort of looks and walks like a human to complete the final steps of a self-driving delivery from vehicle to doorstep.
Outfitted with a lidar and a few stereo cameras, Digit has enough sensory power to navigate through basic scenarios. If it encounters an unexpected obstacle, it can send an image back to the vehicle, which can configure a way round it.
The US Postal Service alone delivered more than six billion packages in 2018, or double the volume it was handling about ten years ago. To help address this issue, Ford is teaming up with Oregon-based Agility Robotics to explore a new frontier in the world of autonomy.
Together, the two companies are working towards making sure self-driving vehicles are outfitted to carry out the final step of getting a delivery from the car to the door.
“Since self-driving vehicles can potentially move people and goods simultaneously, they hold great potential to make deliveries even more convenient and efficient,” said Ken Washington, chief technology officer at Ford Research & Advanced Engineering. “A ride-hailing trip could double as a delivery service, dropping off packages in between transporting passengers. And as we’ve learned in our pilot programmes, it’s not always convenient for people to leave their homes to retrieve packages or for businesses to run their own delivery services. If we can free people up to focus less on the logistics of making deliveries, they can turn their time and effort to things that really need their attention.”
Digit was designed and built by Agility Robotics out of lightweight material. It can lift packages that weigh up to 18kg, can go up and down stairs, walk naturally through uneven terrain, and even react to things such as being bumped without losing its balance and falling over.
“As humans, we take these abilities for granted, but they become extremely important when engineering a robot to navigate the nuances of various environments,” said Washington. “Gaining access to a customer’s door often requires walking through obstacles, including going up stairs and dealing with other challenges, which can be hard for robots with wheels to do since only about one per cent of homes in the USA are wheelchair-accessible.”
Digit has been designed to walk upright without wasting energy, so it has no issue traversing the same types of environments most people do every day.
Its design also allows it to fold itself up tightly for easy storage in the back of a self-driving vehicle until it’s called into action. Once a self-driving car arrives at its destination, Digit can be deployed to grab a package from the vehicle and carry out the final step in the delivery process.
While Digit needs to function on its own, the desire to keep it lightweight and capable of dynamic movement led to an innovative idea. It can tap the resources of another robot – one that’s equipped with advanced sensors and heavy computing hardware – for additional support and analytical skills when needed.
A self-driving vehicle can create a detailed map of the surrounding environment and share those data with Digit instead of having it recreate the same type of information, After all, both Digit and the self-driving car need to know where they are in the world, where they need to go and how to get there.
When a self-driving vehicle brings Digit to its final destination, the vehicle can wirelessly deliver all the information it needs, including the best pathway to the front door. Through this data exchange, Digit can work collaboratively with a vehicle to situate itself and begin making its delivery.
Outfitted with a lidar and a few stereo cameras, Digit has just enough sensory power to navigate through basic scenarios. If it encounters an unexpected obstacle, it can send an image back to the vehicle and have the vehicle configure a way round. The car could even send that information into the cloud and request help from other systems to enable Digit to navigate, providing multiple levels of assistance that help keep the robot light and nimble.
Digit’s light weight also helps ensure it has a long run time, which is essential for a self-driving delivery business that will be operating most of the day.
“Whether we are working side-by-side with robots in our numerous factories around the world or living with them as they help push packages to our door, our primary goal is to ensure they are safe, reliable and capable of working alongside people in intelligent ways,” said Washington. “Through our collaboration with Agility, we are striving to determine the best way for our self-driving vehicles to cooperate with Digit and understand how this new delivery method can be taken advantage of in the future.”