Starship Delivery Robots know where they are going Author: Joan Lääne | Starship Technologies
(plus how to make your own 1: 8 scale papercraft robot model)
By: Joan Lääne, Mapping Specialist, Starship Technologies
Every September when the new school year begins, many first-year students are a little scared of the unknown. Not only about starting school and meeting new people, but also about the journey they have to make every day. They need to learn and remember how to navigate the world and the path to the classroom and the classroom themselves. This can make it easier for a parent to help their child get to know the road better on their first trip here and there, usually indicating some interesting landmarks along the way, such as tall or light buildings or road signs. . In the end it will be trivial for the child to go to school and remember the path. The child will complete a mental map of the world and how to navigate it.
Starship Technologies offers a convenient last mile delivery service with a fleet of sidewalk delivery robots navigating the world every day. Our robots have made over 100,000 deliveries. To get from point A to point B, the robots have to plan a route, which requires a kind of map. Although there are many map systems that are already publicly available, such as Google Maps and OpenStreetMaps, they are limited in that they are designed with car navigation in mind and are primarily focused on car road mapping. Since these delivery robots travel on sidewalks, traveling on sidewalks is safe and requires a detailed map of where to cross the streets as a child needs a mental map to know how to get to school safely and on time every day. So how is this map created?
The first step in creating a map for sending robots is to explore the area of interest and create a preliminary map (2D map) over the satellite images, along simple interconnected lines depicting sidewalks (green), walkways (red), and paths. (purple) as shown in the figure below.
The system treats this map as a graphical node and can be used to create a route from point A to point B. The system can identify the shortest and safest path to be taken by the robot and also calculate the distance and time it would take. to drive this path. The advantage of this process is that it can be done remotely before any robot can physically reach the site.
The next step is to show the robots what the world is like. Similar to the analogy between parents and children, robots need to take a hand in the first time they explore an area. When the robot is first driven, many cameras and robot sensors collect data about the surrounding world. These include thousands of lines coming from perceiving the edges of different features, such as buildings, streetlight poles, and roofs. The server can then create an offline 3D world map from these lines, which the robot can use. As a child, the robot now has a model of the world with guide poles and can understand where it is at any given moment.
Since our robots need to cover different areas at once to complete all their deliveries, in order to be effective they need to put together several maps to create a unified 3D map of a particular area. A unified map is created piece by piece that processes different pieces in the new area until it finally looks like a giant puzzle made up of maps. The server will collect this map based on the line data previously collected by the robot. For example, if the same roof is detected by two robots, then the software represents how it connects to the rest of the map. Each colored line in the image below represents only one part of a trip added to the map.
The final step in the mapping process is to calculate exactly where and how wide the sidewalk is before the robots can drive completely autonomously. This processes the camera images recorded while the robot crawls the field as a reference, as well as incorporating a pre-created 2D map based on satellite images.
In this process more details are added to the map to accurately define the safe places for robots to drive.
Of course, the world around us is not static. There are daily and seasonal changes in landscapes, construction and renovations that change the way the world looks. How can this affect the areas mapped to robots? In fact, the robot’s software handles small and medium changes in the mapped area quite well. 3D models are quite robust and full of large amounts of data, where a tree cut down here or a building thrown there usually does not challenge the robot’s ability to locate its position or use the map. In addition, as the robot drives daily, it continues to collect more data that is used to update 3D maps over time. But if an area is completely remodeled, or new sidewalks are built, then the solution is simple. The map must be updated using new data collected by a robot. Then other robots will be able to drive autonomously again in the same area, as if nothing had happened. Keeping maps up to date is key to continuing to drive robots safely and autonomously.
As you can definitely tell by now, I really enjoy playing with the concepts of 3 dimensional space. Ever since I played 3D first person shooter computer game (Wolfenstein 3D), the 3D world of digital domain has become my interest. I wanted to make my own 3D worlds for computer games, so I found ways to edit existing game levels. Later, I also tried 3D modeling with a computer, which I found interesting. Along with the disclosure and bargain of 3D printers, I also started physically printing models. But much earlier, during the summer breaks at school, I really enjoyed making paper models of different buildings and vehicles. It was an easy and inexpensive way to create something with my hands, but it was interesting to see how a 2D design on a piece of paper, slightly cut, folded and glued, can be turned into a 3D model. Basically, creating a paper craft from a 3D object or “deployment” is somehow the opposite of the map. A 2D design of the surface of a 3D object is being created.
Since I have a passion for paper, I decided to create one for our Starship delivery robots. The goal of making this model is to allow others who can enjoy the same passion that I do to create our own version of our delivery robots. Creating a paper pattern is a fun challenge, and once done, it also becomes a nice decoration. As with creating 3D maps for a robot, making a paper model requires precision, accuracy, and spatial thinking to know how all the pieces fit together. A little patience too.
I have created some instructions for you to create a robot to send your paper craft and I would love to see your efforts. Have fun and good luck making your delivery robot paper model!
Please post a photo of your robot on Instagram and tag it so you can find @StarshipRobots!
Please find the Starship delivery robot papercraft model and instructions here
© Starship Technologies. The design of the Starship® shipping robot and the aspect of the described technologies are proprietary and are protected by copyright and other intellectual property laws.