Transcript: So, what we’ve got for you today is a simulation of a warehouse environment that uses the UnReal Physics Engine to make a realistic representation of this space in the real-world.
Here, we’ve loaded Cyngn’s DriveMod, autonomous vehicle technology, into three warehouse vehicles in this simulation: blue, yellow, and red.
The views of each of these vehicles are represented in the three boxes at the top of your screen. They will give you a sense of how DriveMod processes the world around it.
By the way, none of this has been animated in advance so as we move the avatar around the space, DriveMod will react to the Avatar and the rest of the environment in real-time.
In this first scenario, we have 5 objects in the hallway: three vehicles, one person, and a pallet in the middle of the road.
Now watch as the yellow vehicle dynamically detects the pallet and the avatar and then safely charts a course around these two objects.
Did you see that? Let’s watch again.
Now, in the next scenario, the avatar will come around the corner and encounter another vehicle. If you look in the upper-left hand box, you can see that the blue vehicle has perceived the avatar and calculated a red fence that indicates a “stop” decision. The blue and yellow vehicles both come to a stop to allow the avatar to safely pass.
Now, in this scenario, we want to show you how DriveMod can not only stop for pedestrians but also resolve potential conflicts on its own. Here, the avatar crosses in front of the blue vehicle, and as soon as it is out of the way, the blue and yellow vehicles both speed by.
Let’s look at that in a similar scenario. The Avatar is going to come around the corner and boom, the blue vehicle detects it instantly and comes to a stop. As soon as the Avatar is out of the way, both the blue and yellow vehicles continue on.
Switching views, if you look at the blue box above, you can see much more clearly how the vehicle perceives the world.
Here, the blue vehicle detects a small object on the side of the path. In the blue box, you can see that the object gets flagged with a green fence. In our system, the fences represent decision points where the vehicle must determine for itself whether it is safe to continue.
Same thing here. The vehicle observes the obstacle in the road, our system flags it with a yellow warning, and then immediately charts a new path around it.
Cyngn brings autonomous prediction, planning, and control to all kinds of vehicles across many different environments — from the inside of a warehouse to the bottom of a mine.
This video is just one example that demonstrates how DriveMod’s advanced software enables autonomous vehicles to navigate the world
