SLAM Use Cases. Part 3. Precise Autonomous Drone Delivery without GPS


In the previous articles, I talked about how to use our platform (SLAM SDK and Navigation SDK) for autonomous flying drones in GPS-denied premises (warehouses, construction sites, etc.) and for cleaning tasks done by robotic vacuum cleaners.

Now I want to talk about how unmanned aerial vehicles (UAV) working on our technology could use standard sensors for precise autonomous delivery with 1-inch accuracy.

You can ask why GPS is not suitable for delivery purposes? The reason is that the accuracy of GPS is about 6 meters, so drone working on GPS can deliver goods close to the house, but it can’t autonomously land the parcel on the rug in front of the door. A drone will either slams into the door or does not reach it at all – the accuracy of the GPS system does not allow it to do so.

Screen Shot 2016-09-02 at 11.36.58 AM

What to do?

In this case, it is necessary to use a visual sensor that every drone has – a camera.

Our platform (SLAM SDK and Navigation SDK) can solve the problem of the last “20 feet” of delivery with an accuracy of 1 inch.

Here is how a drone with our installed platform works:

  1. Using the standard function of autopilot on GPS the drone has the ability to launch itself from the starting point and ascend to a safe cruising altitude and then navigate to the desired delivery point—say, a house’s front stoop.

slam gps-denied

2. When the GPS delivery destination is reached, the drone switches to our proprietary SLAM technology, detects and analyses the place where it should land using standard RGB camera and IMU sensors, then descents, and executes the package drop-off.


3. Then the UAV can ascend back to a safe cruising altitude, switch GPS back and navigate safely back or to a new assigned location.

slam sdk in gps-denied environment

Looks simple, isn’t it?

Many global companies that specialise in delivery (DHL, USPS, Amazon, etc.) have been testing intensely UAV to transport packages, food or other goods. A few even launched special programmes with a goal to move delivery to a new level. The bottleneck of all these systems is that they can not realize a complete delivery cycle – there is a need for a certified pilot for each delivery operation. It means that the company bears the costs comparable to the current models of delivery.

By combining GPS and our platform (SLAM SDK and Navigation SDK) the completely autonomous delivery can be ensured and the cost of one package delivery would be ten times smaller.

I am more than sure that in 2-5 years such systems will become widespread and become a part of our everyday life.

Vitaliy Goncharuk
CEO of Augmented Pixels

Leave a Reply