Innovators in the evolving automotive ecosystem converged at the recent Autotech Council meeting, hosted by Western Digital, to share their visions for a self-driving future. What their prototypes and solutions had in common was a shift toward processing at the edge and the use of artificial intelligence (AI) and machine learning to enable an autonomous future.
Read on to hear from automotive innovator Udelv, a startup bringing groceries to your doorstep with autonomous delivery with AI and machine learning.
Does the future of driving mean groceries delivered autonomously to your driveway?
Move over, on-demand food-delivery services. There’s a new delivery company in Silicon Valley. And it just might put an end to those tedious runs to the local grocery store.
Imagine a world where your groceries are delivered to your driveway. It’s a ten-second walk to your driveway instead of a ten-minute drive to the store. Food that’s delivered, not by someone that you don’t know, but by… well, no one at all.
We’re talking about the rise of autonomous delivery, and it’s not as far away as you might think.
A New Way of Food Transportation
You might think that autonomous delivery is as simple as filling a self-driving car with groceries. But, there’s many more pieces to the puzzle. Just ask Daniel Laury, Chief Executive Officer at Udelv. His startup is transforming the way that delivery is done, starting with markets and restaurants.
To make this happen, Daniel and his team have worked with their customers to organize the cargo space on self-driving cars to get the most payload possible. After all, packing more groceries on each vehicle’s trip could reduce the cost per mile. The result? An impressive mix of design, mechanical engineering, and ease of use.
Another issue that comes up with autonomous delivery is road navigation in tight spaces. Just think of a busy parking lot. To maneuver in such a hectic environment, Udelv has focused on high precision. Getting within a few feet is too far. One-to-two centimeter accuracy is the goal.
The All-Important First and Last Mile
On busy roads, driving conditions are complex and constantly changing. For a human driver, there are two main issues to deal with. The first problem is identifying and understanding different scenarios. Should you brake or accelerate? Turn left or right?
And then, there’s the actual driving decision. The amount of time after you make a decision in your head to physically perform a driving maneuver in your car. This latency can mean the difference between a close call and a full-blown automobile accident. That’s why autonomous vehicles need to store, process and analyze data close to the source, at the edge. Technologies such as artificial intelligence might offer autonomous vehicles the ability to understand different driving scenarios and make real-time decisions.
The Role of Data to Make Autonomous Delivery a Reality
What enables this autonomous delivery to happen? Today, real-time processing happens at the edge because there’s simply no time for latency! Data aggregation, big data analytics, model training, and near real-time processing can happen on the network or in the cloud, however, there are challenges and limitations at each juncture. The car is subject to environment, design and cost limitations; whereas the network and cloud have limitations in bandwidth, connectivity, latency and security. As a result, edge devices including cars and all of their sensors and systems, need to be equipped with enough compute power to support these new use cases and storage to keep all the data local.
Western Digital is committed to the automotive market. Data is evolving and growing at an insatiable rate driven by technologies such as IoT, ADAS, cutting-edge infotainment and safety systems, and other connected technologies. A new approach to data storage is required to deliver speed, agility and longevity for various applications, workloads and outcomes that can make it economical to make data alive at scale. We’re driving the innovation across every layer of the infrastructure necessary to stay ahead of new demands. Our breadth of expertise and level of integration give us an unmatched ability to deliver carefully calibrated solutions for every type and use of data.