CHANNELS

Aerospace

Automotive

Automated & Autonomous Vehicles

Battery & Electrification Technology

Defense

Drone TV

Electronics

Manufacturing & Prototyping

Materials

Medical

Motion Control

RF & Microwave Technology

Robotics & Automation

Sensors & IoT

Test & Measurement

Aerospace
Automotive
Automated & Autonomous Vehicles
Battery & Electrification Technology
Defense
Drone TV
Electronics
Manufacturing & Prototyping
Materials
Medical
Motion Control
RF & Microwave Technology
Robotics & Automation
Sensors & IoT
Test & Measurement
Aerospace & Defense
Search Results

Designing a Better — and Safer — Future for Transportation

What can a 20-by-20-foot “smart” scaled city and a fleet of small, motorized cars, drones, cameras, and virtual reality technology tell us about the future of transportation? A lot! Watch this video from Cornell to learn more.

“If you don’t have an experimental testbed, you use simulation. And simulations are doomed to succeed. They’re always perfect,” said the lab’s director, Andreas Malikopoulos  , professor of civil and environmental engineering in Cornell Engineering. “But in the real environment, you have miscommunication, errors, delays, unexpected events. This testbed can give us the opportunity to collect data and extrapolate information, something that we couldn’t do in the real world with real cars, because of safety concerns and the need for resources and space.”

Topics:
Automotive Transportation Unmanned Air Vehicles/Systems (UAV/UAS)

More From SAE Media Group

    Transcript

    00:00:04 I mean, if we imagine that all the cars were connected, automated, I think, there are solutions out there and, robust solutions. But when we put the human in the loop, then we create the source of uncertainty. And this will make things more complicated. We're here in the Information and Decision Science Lab. In the IDS Lab, we develop data driven system approaches for cyber physical systems, systems that they have a cyber component in the physical entity. And we try to make these cyber physical systems able to realize the optimal operation

    00:00:41 while interacting with the environment. So with this scale city that we have built, this is 20 by 20 feet or six by six meter long. We have about 75 robotic cars and about 15 drones. And we try to create the test bed that can allow us to do experiments with, within the transportation environment, without having concerns about safety, or having to spend a lot of resources. With this testbed we can do experiments repeatedly, collect data, and draw concluding remarks based on what we have seen in this scale environment. We can

    00:01:23 purposely having bad actors in the testbed and then design controllers or, algorithms that can help autonomous vehicles, address this bad behavior. We do have the capability to drive these cars, and then we have a mixed traffic environment with human driven cars and connected automated vehicles, CADs. And then we try to develop, you know, the theory algorithms that can make, these vehicles coexist in a safe manner. We usually consider the drone and the cars working together. So the main goal is to demonstrate that we can coordinate different

    00:02:05 type systems together. Like if you have different vehicles, different system, you have different dynamics for them. And they need to exchange some sort of informations and is not simple in real world. Eventually using this data through the experiment testing our algorithm and we want to use this in the real world to solve the traffic problems. You know, it's still a gap between the current situation and the fully autonomous vehicle cars.

    00:02:38 So we also have to think about in the middle time, how can you solve the human driving vehicle cars with autonomous vehicle cars? Large language model now is really hot topic because it has ability to reasoning and give user suggestions, decisions. And we are doing the control. Definitely we don't want a large language model to take over the people with the decision, but we want the large language model can help us. For example, give us some suggestions to analyze the environment, provide more information and also maybe some hidden factors

    00:03:10 the human cannot find, but they can find. Once you come up with a new method, a new algorithm, a new theory, you can prove the concept in simulation. But simulation is doomed to succeed. You need the hardware. We need to go beyond simulation. And this testbed can help into this direction to, help us prove the concept. Validate our theory. Give us the opportunity to collect data, extrapolate, you know, information about this.

    00:03:44 So this this testbed can be in between simulation and the physical world once we do experiments, and once we mature the technology here, then we can move on and do experiments in a real world setting. I think we're getting there. It might take a little bit of time. I'm very optimistic. But, at the same time, we need to make careful steps towards this goal. We don't have to rush this, technologies have to become mature. We need not only technologies, but also new

    00:04:17 generation of engineers that can, you know, understand how we can get to this point and this testbed can help train these engineers and create new knowledge that can get us there eventually.