Making sense out of Sensor Data: DIBOTICS

‘The company provides a suite of Intellectual Property, Algorithms and Engineering services to achieve mobile autonomy, advanced 3D perception and situation awareness.’

Based in Paris, DIBOTICS was created in 2015 by Raul Bravo and Olivier Garcia in order to extend to new application fields the result of 15 years of world-class know-how and field experience on GPS-denied and infrastructure less navigation.

The company’s technology is used as a core component of some of the most advanced products and companies in mobile robotics, operating in both indoor and outdoor environments. The company has developed and field-validated a unique 6 Degree of Freedom Sensor-Agnostic Localization technology, relying exclusively on the data originated from a single sensor: no odometry, IMU or multi-sensor fusion are needed. It works exclusively with OEMs through long-term licensing and knowledge transfer agreements, helping them to enhance their current tele-operation capabilities and develop advancements in vehicle automation and 3D Perception.

In Conversation with the Key Executive

Why was the company set up? How did you select the vertical and decide to be a part of the global platform?

Something has changed in the market relatively recently: a new generation of sensors have appeared, allowing for a native 3D point-cloud output. These sensors include 3D LiDAR (Velodyne, Quanergy, ASC) but also ToF Cameras (Softkinetic, Realsense). We believe that this new kind of data output is a key enabler of a new kind of real-time processing software that is able to take advantage of the potentially millions of points of data per second that can be generated.

Using only the data from a LiDAR, we’re able, simultaneously and in real-time, to:

1. Localize a 3D sensor in 6 degrees of freedom (X, Y, Z, pitch, yaw, roll), without GPS, without Odometry or IMU
2. HD 3D mapping: creation and update, Point-cloud or Vector data
3. Detecting &Tracking moving objects
4. Segmentation of the surroundings (roads/lines)
5. Fusion multiple Lidar data without needing Calibration and without needing Synchronization

Tell us about your first product that was launched.

We’ve developed a technology aiming specifically at the Autonomous Car, called “Augmented Lidar”, allowing you to fulfill most of the features you need to create a Self-driving car from scratch:

• Knowing where the car is when GPS is not available or not reliable enough, called Localization
• Understanding the surroundings in 3D: 3D Mapping, Creation of a map and understanding changes from last perception
• Detecting and tracking objects around the car
• Fuse several sensors

How successful was your first project roll on? Share the experience.

We’re currently on private beta-test with some of the most significant OEMs, Tiers1 and key players on the Autonomous Car scene. The first conclusion is that our approach is a totally new way of dealing with 3D data in real-time, using an off-the-shelf Intel board (our processing of the LiDAR data doesn’t require a GPU). 

What kind of mixed responses have you received from your consumers over the years? How have they motivated you to shape your offerings/grow the company?

The first reaction is always skepticism, as our approach requires a really different way of thinking about the same kind of problems in order to find original solutions.

Because of these skeptical reactions, when having a first meeting with a prospect, today we always start the meeting with a live demo of the solution running on a standard Rental Car (no specific setting is needed).   If for some reason our customers or prospects are not able to see this live demo, they keep being skeptical until they actually see the technology running.

These first reactions were motivating us because that meant that people were not really working in the same way, using the same approach, so we could add value with our technology.

What were the grounds on which you have expanded your company and its offerings over the years?

As we had the experience of creating other companies in the mobile robotics fields (warehousing, industrial floor-cleaning, security), we started commercializing the technology to other companies willing to create new kind of mobile robots. 

Only recently we decided to enter the Autonomous Car market, as this is all about mobile robotics, in fact, so we could apply our experience to this field in a successful way.

What learnings have helped you to set this company up?

We’re a serial entrepreneur team, having worked mainly in mobile robotics and cumulated more than 300,000 real-life working hours of professional products using our technology in challenging applications. This real-life, field-based experience has shown us the gap between theory and practice; between a prototype and a working product.

Do you have any new products ready to be/getting ready to be rolled out into the market?

Yes, we’ll be launching a new product in two months for Autonomous Driving that will go well beyond the state of the art. 

Where do you see DIBOTICS a couple of years from now?

We see our technology as a key component running in real-time in most mobile robotics applications, not only Autonomous Car, but any mobile robot, Drone and even VR devices.

Meet the Master

Raul Bravo, CEO and Co-Founder: Raul and his Co-founder Olivier Garcia, are the creators of the Augmented Lidar technology. He is a serial entrepreneur and start-up coach, with an extensive 15 years of background in both bootstrapped and VC-backed start-up creation and growth.

Raul holds a Degree in Engineering from UPC (Barcelona, Spain) with an MBA from College des Ingénieurs (Paris, France). He has filed 10 patents and obtained 27 different awards for his engineering and entrepreneurial career, among them the MIT Technology Review – Top 10 Innovators Under 35 in France.

“The DIBOTICS technology is used as a core component of some of the most advanced products and companies in mobile robotics, operating in both indoor and outdoor environments.”