"For any project that is truly a breakthrough

[tanjimajuha20]

Russian developer of visual-inertial mapping, localization and navigation technologies, deeptech startup Dioram (Skolkovo resident and portfolio company Skolkovo Ventures) has begun integrating the Dioram Visual Navigation location system with the "digital twin" project of Moscow roads by Digital Roads, a Skolkovo resident. In 2018 and 2020, Dioram attracted about $1 million in investments from the Skolkovo Ventures Business Angels Club.

The cooperation contract india cell phone number list between the two companies, signed in September 2022, opened up new avenues for the development of Smart Cities projects without relying on insufficiently accurate and reliable location technologies such as GPS. Diorama's proprietary and entirely Russian SLAM (simultaneous localization and mapping) technology, which has been developed for over four years, has opened up new opportunities for domestic projects in the field of unmanned transport, robotics, AR/VR and "digital twins".

in scientific development, the most difficult stage is the transition from laboratory prototypes to real-life applications. Over the past four years, we have been developing the "core" of the Dioram SLAM One technology and were ready for commercial implementation at the beginning of 2022. In addition to autonomous transport, robotics or AR/VR, the ability to perform high-precision navigation in a city without GPS opens the way to a large number of potential solutions, including for Smart Cities," commented Vasily Meshcheryakov, CEO of Dioram.

For the Autodiscovery system to function, it is necessary to know its exact location at every moment of the complex's passage. GPS technology, which appeared at the end of the last century, despite its widespread use, is no longer able to meet modern challenges for the development of unmanned transport, mobile robots or smart cities. Even with the use of RTK ground-based base stations, satellite navigation still has insufficient stability and coverage in urban areas, low accuracy in places or paid subscription requirements. It is also worth noting that GPS/RTK is unable to solve the problem of determining location indoors.

"The Autodiscovery system allows you to create digital models of road infrastructure, track all changes that occur, control the quality of road maintenance, and identify deviations that are dangerous to road safety. Thanks to the developments of Dioramas, carried out on our order, we plan to provide the market with compact devices for navigating drivers and autopilots using our high-precision digital maps," said Andrey Borodin, CEO of Digital Roads.

The developments of "Dioramas" are based on computer vision and machine learning technologies and are aimed at developing their own SLAM algorithms. SLAM technology solves the problem of simultaneous localization of a device in space without external sensors and building a map of a three-dimensional environment (SLAM = Simultaneous Localization and Mapping).

For a joint project with Digital Roads, Dioram research engineers are adapting the company's existing developments to eliminate GPS or similar navigation tools. First, the Dioram SLAM One algorithm uses panoramic cameras installed on the Digital Roads Autodiscovery complex to create "digital maps" of Moscow in the form of point clouds. Then, by mixing data from the cameras with readings from inertial sensors, the algorithm "recognizes" itself on a previously created map. It also builds trajectories in a completely unfamiliar area if necessary. Thus, it is possible to determine the location of the Autodiscovery automobile complex in real time with an accuracy of up to 10 cm and without using GPS or similar technologies. Using data from the cameras on the car, an accurate trajectory of movement and a three-dimensional "map" of a Moscow street from a point cloud are formed, completely coinciding with the contours of objects on the satellite image.