US e-commerce giant Amazon is set to deploy a new AI-powered solution, Vision-Assisted Package Retrieval (VAPR), on 1,000 electric delivery vans by early 2025.
The technology is designed to assist drivers in locating the correct packages for delivery quickly and efficiently, using a system of visual and audio cues.
Go deeper with GlobalData
VAPR will be integrated into Amazon’s electric delivery vans produced by Rivian, aiming to save drivers time and effort.
The concept of VAPR originated at the beginning of 2020, when the retailer’s transportation team envisioned using technology to aid drivers seamlessly.
After several years of development and driver feedback, the tech is now ready for a wider rollout.
Amazon transportation product manager John Colucci said: “We had to think about factors that are unique to the delivery experience, such as lighting and space constraints inside vans.”
How well do you really know your competitors?
Access the most comprehensive Company Profiles on the market, powered by GlobalData. Save hours of research. Gain competitive edge.
Thank you!
Your download email will arrive shortly
Not ready to buy yet? Download a free sample
We are confident about the unique quality of our Company Profiles. However, we want you to make the most beneficial decision for your business, so we offer a free sample that you can download by submitting the below form
By GlobalDataOnce a delivery van arrives at the delivery location, VAPR will project a green “O” on all packages for that stop and a red “X” on the rest.
This system allows drivers to quickly identify the correct packages without searching through labels or using mobile devices.
Bloomfield Logistics driver Bobby Garcia, who has been testing VAPR in the North Boston area, said: “Before, it could take me anywhere between two and five minutes to empty a tote and organise packages for the next stops. Now, with VAPR, this whole step just takes me about a minute.”
VAPR employs a computer vision technology, Amazon Robotics Identification, or AR-ID, created for fulfilment centres, to automatically recognise items during inventory stowing or picking. This system eliminates the requirement for manual barcode scanning.
By analysing its environment, it can detect and interpret several barcodes simultaneously in real time.
VAPR’s development involved training machine learning models to recognise labels and packages under different lighting and packaging conditions.
The technology has been optimised for use within a van environment and is paired with automotive light projectors and cameras, all linked to the van’s delivery route navigation system.
VAPR utilises several Amazon Web Services technologies, including the machine learning platform SageMaker and IoT Greengrass.
In September 2024, Amazon announced an additional investment of $2.1bn into its delivery service partner (DSP) programme.
