{"id":45304,"date":"2020-02-12T16:02:43","date_gmt":"2020-02-12T23:02:43","guid":{"rendered":"https:\/\/www.realsenseai.com\/uncategorized-cn\/pudu-delivery-robot\/"},"modified":"2026-03-19T15:53:59","modified_gmt":"2026-03-19T21:53:59","slug":"pudu-delivery-robot","status":"publish","type":"post","link":"https:\/\/www.realsenseai.com\/cn\/news-insights-cn\/pudu-delivery-robot\/","title":{"rendered":"Pudu Robotics and RealSense Technology"},"content":{"rendered":"\nWhen was the last time you visited a restaurant and looked around in vain for your server to attract their attention? This is a fairly common experience for many people \u2013 restaurants get busy, and servers struggle to keep up with the numerous tasks required of them, from taking orders and clearing tables to delivering food. Shenzen based startup Pudu Technology is working at easing some of the burden servers carry, by providing restaurants with highly automated food delivery robots. The robots are not designed to replace human servers, but rather to make it easier for servers to spend more time with their customers, outside the kitchen.\n

What is the experience?\u00a0<\/strong><\/h3>\nThe robots have a touchscreen that allows a food expeditor to assign each tray of the robot (they can have up to 4 trays) to a specific table. Once the table is selected, the robot navigates its way through a crowded restaurant to the table, and will then give instructions to the diners to allow them to take their own plates and food items from the trays \u2013 for example, telling you that your items are on the 3rd<\/sup> and 4th<\/sup> trays. Once you\u2019ve taken your items from the robot, a simple button press on the touchscreen instructs the robot to continue on its journey, either returning home or going to the next table. Meanwhile, your server has a significant amount of time saved \u2013 time that they can now spend with you. By utilizing a robot to assist with a routine and repeatable task, the server can use their time more efficiently on connecting with and serving the needs of their customers.\n\n
\"Closeup
Closeup of the touchscreen interface.<\/figcaption><\/figure>\n

Technical details<\/strong><\/h3>\nThe robots navigate using a proprietary PuduSLAM algorithm which integrates data from multiple sensors; Lidar, vision camera with a sensor pointing directly up, and two \u00ae RealSense\u2122 Depth Cameras located in the lower part of the robot. The depth cameras are crucial for avoiding collisions and adapting to a changing environment. In an average restaurant, you can expect chairs and table legs to move, have bags or coats on the floor or placed on the back of chairs or booths, not to mention people moving around the space. Tablecloths can also present challenging problems to 2D cameras, where depth cameras are more able to deal with all of these different factors and problems with ease.\n\nThe combination of the upward facing camera, VSLAM calculations and the depth cameras allows the robots to navigate these changing environments. Stand in front of the robot, and it will politely either try to navigate its way past you, or it will ask you very politely to move, as it says \u201cI need to work.\u201d Once the way is clear, it will resume its journey to your table, where your food is delivered hot and fresh.\n\n