By Satya Vakkaleri, VP of Product

Image for post
Image for post
Undelivered Boeing 737 Max planes in Seattle, Washington, on August 13, 2019

“If we knew everything back then that we know now, we would have made a different decision,” said Boeing’s ex-CEO Dennis Muilenburg during the senate hearing in the aftermath of the two plane crashes that took the lives of 346 people. Boeing, however, knew its system had a problematic single point of failure two years before Max was cleared to fly. The design of the new flight control system was relying on information from just a single angle of attack (AoA) sensor with no redundant system. Needless to say, this debacle led to the firing of the Boeing CEO soon after. …


By Satya Vakkaleri, VP of Product

Image for post
Image for post
Image courtesy: McKinsey

Opportunity at a time of crisis

Amidst all the gloom surrounding the current pandemic, certain markets are expanding rapidly. Crisis is becoming a catalyst for growth. While market opportunities in video conferencing and tele-medicine are all over the news, the less obvious ones include investments in the autonomous vehicle delivery market that are accelerating the future of autonomy.

Neolix, a maker of urban robo-delivery trucks, has raised $29 million to fund the mass production of its delivery bots. UDI vans have made more than 2,500 autonomous trips, and the company has hired 100 employees to put its assembly line into high gear. …


By Scott Harvey, CTO and Satya Vakkaleri, VP Product

Civil Maps’ localization in San Francisco

To LiDAR or not to LiDAR, that is the question

In the previous blog post, we introduced the concept of using a single sensor for localization. Before jumping into the details of the Civil Maps solution, we would like to take a moment to celebrate the news of Volvo’s next generation of cars to be equipped with LiDAR. It is not surprising that a car manufacturer known to put safety first has bucked the industry trend and decided to use LiDAR.

In the last year, Civil Maps doubled down on its hypothesis that LiDAR would become a mainstream sensor in the future. We continued to actively pursue our product roadmap to build LiDAR only localization. Since then, we have seen a major pull from the delivery market with bots equipped with LiDAR. Now, we have a major OEM making a historic move to include LiDAR in their next generation of cars. While it is too early to make a definitive prediction and celebrate along with the LiDAR manufacturers, we definitely see promising signs in LiDAR’s future. Its acceptance in other non-automotive applications such as AR will only push innovation of the LiDAR sensor further, and simultaneously lower the sensor unit cost in the years to come. …


By Scott Harvey, CTO and Satya Vakkaleri, VP Product

Where am I?

Accurate LiDAR localization in a narrow road flanked by buildings
Accurate LiDAR localization in a narrow road flanked by buildings
Accurate LiDAR localization in a narrow road flanked by buildings

Accurate estimation of an autonomous vehicle’s position and orientation known as “localization” is not just a prerequisite for decision making and path planning, it is critical for the safe operation of the vehicle. Autonomous vehicles are equipped with different types of sensors such as GPS, IMU, Camera, LiDAR, Wheel Encoders, and Radar that can all help localize the vehicle. Some of these sensors are redundant, providing independent measurements of the same property; while some are complementary, and when combined will provide a more complete representation of the phenomenon under observation. Traditionally, arriving at an accurate localization estimate involves sensor fusion — a process that takes data from different sensors and utilizes them to reduce the amount of uncertainty.


Image for post
Image for post

By Nicholas Stanley, Technical Operator, Civil Maps

Maps can be exceptionally powerful — a well-crafted map can mean the difference between arriving on time… or not arriving at all.

We at Civil Maps have been working tirelessly to bring the next generation of high-definition augmented reality maps (“HD Semantic Maps”) to fruition, but in the case of our recent team-building foray into Napa Valley, a world-renowned culinary and viticultural hub, I didn’t need any special map to navigate.

The city of Napa is my hometown, and it just so happens to be a place that many aspire to make a pilgrimage to, even just once in a lifetime; in contrast, up until recently I’ve commuted away from Napa to Civil Maps’ offices around the Bay Area, nearly every day since I began work in 2016. …


Image for post
Image for post

By Paul Drysch, VP of Sales and Business Development

Today’s autonomous vehicle and automated mobility market looks a lot like the early days of the personal computer industry: many players, each building its own proprietary machine: a highly automated vehicle (HAV). When a universal, hardware-agnostic computer operating system (OS) finally came along, it allowed specialized components — video cards, motherboards, modems, hard drives, and etc. — from a variety of manufacturers to work together through a common interface and software to be written once and run across all PCs using the OS. That operating system brought about the personal computer revolution. …


Image for post
Image for post

By Sravan Puttagunta, CEO and Co-founder, Civil Maps

We’ve all seen augmented reality (AR) in shows such as Monday Night Football, in which broadcasters superimpose virtual lines and diagrams on the field, manually marking up video playbacks to highlight key maneuvers and strategies. This helps viewers easily focus on points of interest, rather than looking at everything on the screen. In this case, augmented reality technology provides an instantaneous, mental shortcut to understanding; viewers don’t need to guess where the action is on their TV or what’s going on. The drawn-in overlay provides that direction. When augmented reality is used for autonomous vehicles, there will be a similar benefit; it can help the car quickly figure out where to focus attention as it drives, eliminating the need for the vehicle’s sensors to search everything they see around them just to find the most relevant road signs. …


Image for post
Image for post

By Sravan Puttagunta, CEO and Co-founder of Civil Maps

Civil Maps is excited to announce that we’ve migrated our edge-based HD mapping and localization solution to the Arm® family of processors. Arm is the licensor to the largest ecosystem of automotive grade system-on-chips (SoC) and system-on-modules (SoM), with its chips already found in 85% of automotive electronic control units (ECU) on the road. Our team sees this as a key step towards building a truly scalable platform for self-driving car developers. …


Image for post
Image for post

By Sravan Puttagunta, Co-founder & CEO of Civil Maps

We are delighted to announce key developments that will accommodate and enhance our rapid growth. Vijay Viswanathan and James Dawson will join the company in our San Francisco headquarters, bringing with them a wealth of experience from the automotive and navigational technology fields. Additionally, we’re expanding into the heart of the Silicon Valley with a new office in Palo Alto that will leverage the area’s diverse high-technology talent pool. …


Image for post
Image for post

Ah, everyone’s excitement about going to CES. I never could understand it, until now. With over 190,000 attendees this year, I don’t think any city could have run CES more smoothly, but it was still quite challenging even for Las Vegas. For starters, I wasn’t prepared for the Uber and Lyft prices to be five times the normal amount and for the gridlock traffic around the convention center. Nor was I prepared for the $500/nightly cost of a no-frills hotel room!

Image for post
Image for post

Still, I was thrilled to be at the event. Our engineering team arrived just a few days in advance of the event, which was a testament to the simplicity and speed of our Fingerprint Base Map™ (FBM) creation process; We were able to create a half a dozen different demo route options in a compressed amount of time. The FBM itself is the localization layer in our HD Semantic Map and we were debuting it at CES. Several companies developing autonomous vehicles had already been using our voxel-based fingerprint technology for their mapping and localization. It was time to showcase this on a bigger stage and with dozens of live demos. To prepare for our appointments, we set up headquarters in a large house near the convention center and got busy prepping our Atlas DevKits on a handful of demo cars. After a few days, we were ready to go and the team felt pretty good about the localization performance. Instead of starting with heavy compute and an expensive hardware setup, we were able to achieve centimeter-accurate, real-time localization with significantly reduced hardware, processing power, bandwidth usage, and storage requirements. This is an approach that I’ve admired from the day I started checking out Civil Maps. Now as a part of the team, it has been incredibly interesting to develop the product and put it to the test in front of industry experts from all over the world. …

About

Cognition for Cars

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store