Perciv develops dedicated AI software for radar perception, enabling robust autonomy for vehicles and robots in all weather and lighting conditions. The platform replaces expensive LiDAR systems by leveraging affordable, off-the-shelf radars, often enhanced with camera fusion for improved accuracy. This modular solution provides core capabilities like localization, object detection, classification, and tracking for on-road, off-road, and aerial applications.
Funding
$2.6M raised to dateRaised to date based on public sources. This may differ from the amount the company actually raised and is based only on what is publicly available on the internet.
DOFounders
Product
Problem
Autonomous vehicles and robotics rely on expensive LiDAR sensors for accurate environmental perception, significantly increasing system costs. These systems often struggle in adverse weather conditions, compromising safety and reliability.
Solution
Perciv AI offers AI-driven software solutions that enhance the capabilities of existing radar sensors, providing reliable perception comparable to LiDAR at a lower cost. Their technology enables robust autonomous systems that perform effectively even in challenging weather conditions. By using advanced AI algorithms, Perciv AI unlocks the hidden potential of radar data, making autonomous mobility safer, more affordable, and more accessible across various applications. The software is designed to be sensor and hardware agnostic, integrating seamlessly with existing vehicle and robotics platforms.
Target Audience
The primary target audience includes manufacturers of autonomous vehicles (cars, trucks, tractors), robotics companies, and smart city developers seeking affordable and reliable perception solutions.
Features
- AI-driven radar perception software for enhanced object detection and environmental understanding
- Radar-camera fusion technology for improved accuracy and robustness
- Algorithms optimized for performance in adverse weather conditions such as rain, fog, and snow
- Road user detection from 3D radar cube using CNN-based models
- Occlusion-aware sensor fusion for early pedestrian detection
- Free road estimation using radar data for drivable space detection
- Sensor-agnostic design for compatibility with various radar hardware
- Low compute requirements (less than 2 TOPS) for efficient deployment