In the last two decades, there has been huge growth in the e-commerce sector, driven by big e-commerce companies such as Amazon and Alibaba, across the world. This has led to retailers around the globe looking to expand their operations to include online shopping. This has brought opportunities as well as challenges for delivery service companies. As demand increases among competitors, the delivery service companies have scrambled to offer product delivery as early as possible. These delivery times may vary from a few hours to a couple of days. This has created a huge demand for delivery agents as well.

The use of robots in this sector could have huge potential and could enhance business prospects. For instance, robots could be used in packaging or sorting retail items, delivering items within warehouses, or even delivering the items to our houses. This way, they could be more productive and deliver at least twice as fast as a human could in terms of time and the number of deliveries.

Let's consider this as our use case and use the mobile manipulator we created in Chapter 3, Building an Industrial Mobile Manipulator, to act as a delivery agent. Generally, the products are shipped from the retailers to the specific city and then they're split and sent to different streets that have a common delivery office for every street or group of streets. We shall simulate that delivery office scenario after the products have arrived from the retailer in Gazebo and ROS. Our environment shall look as follows:

Here, the robot would be located in the post office. The robot is supposed to deliver items to the three houses in the colony and are numbered 1, 2, and 3, respectively:

Once the items are received by the post office, the office shall know of the list of products that need to be delivered to specific locations. This information is shared with the robot through a central material movement application software. Furthermore, the products are placed in the delivery area that the robot has access to. The robot then receives the list and updates itself with the delivery locations, along with the respective products, and plans its delivery.

In this case, the robot needs to know the products, know the locations, and pick and deliver the items to and from the necessary locations. If we break the application into chunks, we should make the robot base autonomous, then make our arm pick and place items intelligently, and finally put all of these into a series of actions using state machines. Now, let's set up our environment in Gazebo.