The Future of Road Safety is Affordable and Robust.

AI for Radars

Perciv offers AI based software solutions to unleash the hidden capabilities of radars and enable robust and affordable autonomous systems, with Lidar-like performance, even in adverse weather conditions.

Unleash your Radars with Perciv.ai

State-of-the-art and AI-driven Software

Spinning off from TU Delft’s Intelligent Vehicle group, our team has conducted state-of-the-art research on automotive radars including radar point clouds, radar cubes and custom-designed neural networks for range-doppler data.

Perciv AI is the next step, where we are developing the research into market ready products, together with partners from the automotive industry.

  • CNN Based Road User Detection Using the 3D Radar Cube

    Read here

  • Occlusion-aware sensor fusion for early crossing pedestrian detection.

    Read here

  • Multi-class Road User Detection with 3+1D Radar in the View-of-Delft Dataset

    Read here

  • 3+1D Radar and Camera based Point Cloud Estimation for 3D Road User Detection

    Read here

  • Radars can serve as a reliable modality to detect drivable space around the ego-vehicle

Our Expertise

From sensor selection to the downstream perception task evaluation, Perciv.ai handles it all.

Partners

  • Abz Innovation
  • Yes Delft
  • tu delft logo
  • NWO LOGO
  • Nvidia Inception
  • Delft Enterprises logo
  • provinci Zuid-Holland
  • Horizon2020
  • RvO
  • Robo House logo
  • Philips Innovation Award

News