Overview

Sensor Sim, developed by Applied Intuition, is a real-time sensor simulation platform designed to enhance the development, testing, and validation of advanced driver-assistance systems (ADAS) and automated driving (AD) systems. By providing a virtual environment to simulate sensors such as cameras, lidars, and radars, Sensor Sim enables engineering teams to create diverse and complex scenarios that might be challenging or impossible to replicate in real-world testing. This approach accelerates the development process, improves system safety, and reduces costs associated with physical testing

Features and Capabilities

• Physically Based Rendering: Utilizes multi-spectral rendering techniques to accurately model sensors across the electromagnetic spectrum, including camera, lidar, radar, and ultrasonic sensors. Employs path tracing for high fidelity and hybrid ray tracing for optimal performance, minimizing the domain gap between simulated and real-world data.​
• Real-Time Performance: Optimized for real-time simulation, supporting hardware-in-the-loop (HIL) testing and large-scale software-in-the-loop (SIL) testing. Allows direct configuration to balance performance and fidelity according to specific testing requirements and budgets.​
• Comprehensive Scenario Creation: Enables the development of complex, customizable 3D environments to simulate a wide range of driving conditions and scenarios, including edge cases that are difficult to reproduce in reality.
• Automated Ground Truth Generation: Programmatically generates ground truth labels such as bounding boxes, depth information, semantic segmentation, and optical flow, facilitating efficient training and validation of machine learning-based perception systems.​
• Sensor Suite Design and Validation: Models real-world sensors to test thousands of potential configurations, analyzing performance to identify sensor suites that optimize functionality while minimizing costs.​
• Integration with Digital Twin Technology: Collaborates with partners like Valeo to provide digital twin platforms, offering extensive and diverse scenarios for ADAS perception testing and validation, leading to more accurate simulations.​
• Enhanced Edge Case Testing: Allows the creation and simulation of edge case scenarios, such as pedestrian detection and avoidance in urban areas, ensuring comprehensive testing of ADAS features.​
• Cost-Effective Testing: Shifts a significant portion of real-world sensing and perception development, testing, validation, and data collection into simulation, potentially saving millions annually.​
• Accelerated Time to Market: Facilitates faster development, testing, and validation of sensing and perception systems, shortening the time to deployment without compromising safety or quality.