Product: Safety-Critical LiDAR Perception System for ADAS

Client Challenge:
The project focused on ensuring Functional Safety compliance for a LiDAR-based perception sensor used in Advanced Driver Assistance Systems (ADAS). The objective was to develop and refine safety concepts, perform system and software safety analyses, and establish robust safety mechanisms for perception-related functions in accordance with ISO 26262, while incorporating emerging AI/ML safety guidance.

RSB Solution:

• Supported Functional Safety activities for a safety-critical LiDAR perception system throughout the ISO 26262 development lifecycle.
• Contributed to the development and refinement of Functional Safety Concepts (FSC), Technical Safety Concepts (TSC), and safety concept inputs for software running on a computational platform.
• Performed system-level safety analysis of the LiDAR sensor and software-level safety analysis of perception and detection functionalities.
• Analyzed and refined Functional, Technical, and Software Safety Requirements (TSR, SSR), ensuring consistency with the functional architecture and software implementation assumptions.
• Coordinated and supported SYS FMEA and SW FMEA activities, including failure mode analysis, failure effects evaluation, diagnostic concepts, fault detection, fault reaction, and safety mechanism definition.
• Maintained end-to-end traceability between safety requirements, safety analyses, functional architecture, software assumptions, and verification evidence using Medini Analyze.
• Supported early adoption of ISO 8800, ISO/IEC TR 5469, and SOTIF-related concepts to assess AI/ML safety considerations for automotive perception systems.

Delivered Results:
The project delivered Functional Safety work products supporting LiDAR perception software and system development, including safety concept refinements, system and software safety analyses, SYS FMEA and SW FMEA contributions, diagnostic concepts, safety mechanisms, and traceability between safety requirements, architecture, and safety analyses. The project also established an initial framework for applying AI/ML safety guidance to perception-related automotive functions.

Client Value:
The client gained a robust Functional Safety foundation for a safety-critical LiDAR perception platform, improving compliance with ISO 26262, strengthening safety analysis and traceability across system and software domains, and supporting future integration of AI/ML safety practices for advanced ADAS applications.