Perciv AI: safer roads with smart radar

By Tessa van Breeden - How do self-driving cars and robot vacuums move independently? They use smart cameras and sensors to turn, stop, or go. Despite their affordability and wide availability, radar sensors lacked the advanced perception capabilities of their counterparts, such as lidars and cameras - a gap that Perciv AI's innovative software fills.

The Journey from Research to Revolutionizing Radar Systems

Andras Palffy did his PhD at the TU Delft with the Intelligent Vehicles Group. He focused on enhancing radar-based detection in intelligent vehicles and graduated in February 2022. Srimannarayana Baratam, motivated by the pressing issue of road safety in India,  joined him and finished his master’s thesis guided by Andras. Finally, Balazs Szekeres met Andras via a guest lecture, who recognized the potential of their research. Together, they ventured to bring AI-based software to improve radars to the world by starting Perciv AI. Their solution helps radars to perceive their surroundings and understand the world better. It will advance autonomous robot solutions such as self-driving cars, UAVs, or, for example, delivery robots in smart cities.

Early on, we showed our technology to a potential partner, but they wouldn’t believe it works. We had to swear on our lives, and they did a successful pilot with us.Andras Palffy, co-founder of Perciv AI

Perciv AI’s Impact on the Automotive Industry

The automotive industry is working towards self-driving technology, though fully autonomous cars could take many years. For example, ensuring vehicular safety is vital for this revolutionary road, an area in which Percive AI’s software can help. Existing autonomous systems rely heavily on sensors like lidar and cameras, which require clear weather and lighting conditions to function optimally. However, there wasn’t a similar solution for radar sensors. In contrast, radars are affordable, robust, and have a good perception even in adverse weather conditions. So, Perciv AI took this gap as an opportunity to develop AI-based software that synergizes with radar sensors, effectively making decisions on data gathered by radar sensors. Its software enables radars to interpret and make decisions based on the data they collect, transforming them into intelligent systems capable of enhancing safety in all driving conditions. Though there is much potential for many different applications, such as agricultural drones, lawnmowers, and hospitality robots, Perciv AI decided to focus on the automotive industry first.

Ultimately, we focused on the automotive industry first because we believe our solution will save lives. In the dark or heavy rain, I might not see a cyclist without lights, but the radar in my car will and will prevent an accident.

Srimannarayana Baratam, co-founder of Perciv AI

From Vision to Venture

Perciv AI is graduating from the early startup stage and validating its product. They just hired their first employees and are running several pilots with automotive partners now. They received help from YES!Delft, where they were accepted into the Validation Lab program first and the Accelerator Program after. Through the clear framework taught in the incubator, the network, and the help they received, they managed to create a minimum viable product (MVP) based on the needs of the industry. For that, they won the Rookie of the Year award from YES!Delft in 2023.

We wished we had done the Validation Lab three months earlier; it was beneficial, and we improved our MVP. Also, we are grateful for the Accelerator program. We have a great team now, which helped us understand how to manage our company.

Balazs Szekeres, co-founder of Perciv AI

We are very proud of our current team, consisting of two employees and interns. We are very happy with their dedication and work. Perciv AI could not be here without them.Andras Palffy, co-founder of Perciv AI

Mapping the Future: next steps in the upcoming 6-12 months

Perciv AI is partially bootstrapped, partially funded by grants, and already generating early revenue. They are actively seeking additional funding to propel their radar-enhancing software into the automotive market more rapidly. The team is in the final stages of their proof of concept projects, analyzing data and refining the software to customer needs to ensure seamless integration and optimal functionality. Perciv AI is ready to use its advanced AI solutions to enhance existing car radar systems, boosting road safety worldwide.

Your dream can become a reality too!

As long as you want to make a change, don’t miss the opportunity to boost your startup by joining the YES!Delft Validation Lab or Accelerator Program. However, if working for a startup resonates more with your ambition, we got you! Check out vacancies placed by our startups.

Be the change, as change starts with you.

To top