iHLS Accelerator Startup: Breakthrough in GPS-Denied Autonomous Flight
This post is also available in: 
עברית (Hebrew)
The world of autonomous flight opens up numerous possibilities and applications, as drones will be able to fly autonomously from one place to another and supply valuable, real-time information or provide deliveries. However, this technology that is about to change our lives in the near future will have to solve some complex problems, especially when it comes to flying above people or vital assets. There is a need for safe and reliable flight also when connectivity or GPS are not available.
An innovative Israeli startup has developed a ground-breaking solution for this challenge in the autonomous realm. Sightec is a startup company, focused on computer vision & AI solutions for autonomous flights. The company develops ‘Visual Situation Awareness’ systems for drones and UAVs that imitate the actions of a human pilot. The company’s current offering includes vision-based navigation (GPS-free) and ‘Wide Area Persistent Surveillance’ software for drones and UAVs.
The company had originally started off as a research project by Prof. Joseph Francos, Chief Scientist and Co-Founder, at the Dept of Electrical and Computer Engineering at Ben-Gurion University of the Negev when three years ago, Roy Shmuel, CEO and Co-Founder, had joined forces – shifting the research to a product company. The two entrepreneurs identified the fact that the autonomous aerial vehicle market was constantly growing and there was a need for a technology that could imitate a human pilot’s ability to safely navigate and locate areas of interest even when GPS is not available or accurate.
The company’s products integrate algorithmics in leading academic research, including developments from Prof. Joseph Francos’ lab at BGU. The company has filed several patents in the image processing fields.
The innovative startup participates in the seventh batch of the iHLS Security Accelerator. According to Roy Shmuel, the program helped them clarify the market needs and understand how to optimize their service via customers’ needs. Thanks to the program, the company gained the vital access to strategic players in the ecosystem.
The company developed a real-time machine vision software using mono-camera as the main sensor running on off-the-shelf edge devices. “We are operating in the realms of aerial autonomy – developing artificial intelligence that imitates the pilot’s operations and provides a safe flight,” said Roy Shmuel. Their technology imitates the capability of the human eye to understand what it is seeing thanks to their software integrated into the camera. “We’re talking about end-unit processing integrating a small, cheap camera,” he added. “The technology enables us to take a real-time video feed and provide insights regarding any scene, including self-location, understanding objects in the environment, and understanding their location in the space.”
Sightec’s software solution for navigation without GPS provides visual navigation with better accuracy than GPS, crash site selection and auto safe landing, as well as real-time detection and tracking capabilities of aerial and ground objects.
The system on the drone identifies and categorizes objects in the environment (people, cars, other drones) and supplies real-time data.
The demand for Sightec’s solution is in the intelligence, first responders and security markets, as well as in the growing drone delivery market.
The 12-people team of the Tel-Aviv-based startup includes experts in computer vision, AI, autonomous systems, etc., some of them had worked in Mobileye in the past.
More than a year ago, the company completed a $2 million seed round, and is now heading to a round A funding. The company’s main investors are Sherpa InnoVentures and Playfair Capital.
The company strives to turn its software into the global industry-standard software for safe and accurate autonomous aerial missions. “Our vision is to help improve flight and flight operation safety with an affordable off the shelf camera as well as to improve the drone performance in critical real-time missions” concludes Roy Shmuel.
