Apollo is a comprehensive autonomous driving platform developed by Baidu, providing a full software stack for self-driving vehicles. The repository contains a modular architecture built on Cyber RT, a high-performance runtime framework optimized for autonomous driving scenarios with low latency and high throughput. Key modules include perception (camera, LiDAR, radar, sensor fusion), planning (scenario-based with multiple planners), control (PID, LQR, MPC controllers), localization (RTK, NDT, MSF), prediction, routing, and V2X communication. The system integrates with hardware through extensive driver support for various sensors (Velodyne, Hesai, Continental radar, Mobileye, etc.) and vehicle platforms (Lincoln, Lexus, Geely, etc.). Dreamview provides a web-based visualization and monitoring interface. The codebase demonstrates strong engineering practices with clear module boundaries, extensive configuration management, comprehensive testing infrastructure, and support for both x86_64 and ARM64 (Orin) platforms. Recent versions introduce package management for easier deployment, Dreamview Plus for enhanced developer experience, and cloud-based model training services (Apollo Fuel).
This repository is a curated suite of 152 Claude AI skills (76 builder + 76 paired checklist reviewer archives) for generating structured Excel workbooks across the automotive engineering lifecycle. Domains include ISO 26262 functional safety, ISO/SAE 21434 cybersecurity, ISO 21448 SOTIF, ASPICE, AUTOSAR, SysML/MBSE, diagnostics (UDS, DBC, ODX), calibration (A2L), quality tools (APQP, PPAP, FMEA), and program management. Each `.skill` file is a ZIP archive containing SKILL.md, Python generators/reviewers, references, and shared utilities. Builders produce multi-tab xlsx workbooks from JSON input; reviewers validate existing workbooks and emit compliance checklists with dashboards. The suite is maintained via an autonomous weekly cycle (plan, polish, docs, release), with status tracking, chain-contract audits, and known issue catalogs. While the concept is robust, many shipped skills suffer from documentation/code drift (advertised check counts mismatch actuals), silent failures (empty workbooks certified as compliant), and duplicated shared scripts across 152 archives.
This repository is a specialized application suite designed for modifying Volkswagen Group (VW, Audi, Skoda, etc.) MIB (Modular Infotainment Architecture) infotainment head units. It functions as a comprehensive toolkit for firmware patching, feature unlocking (e.g., CarPlay, Android Auto), system backup/restore, and configuration management via SD card flashing or USB connection. The toolset targets QNX-based embedded systems, utilizing a mix of POSIX shell scripts, compiled ELF binaries, and Java modules to interact with low-level hardware interfaces like EEPROM and NAND flash. Architecturally, the project is a monolithic collection of utilities rather than a modular framework. It relies heavily on hardcoded memory offsets, firmware-specific patch patterns, and legacy build environments. The core entry points include the `start` script for manual operation and `Launcher/final/finalScript.sh` for automated SD updates. Data flows involve reading FEC (Feature Enablement Code) lists, generating binary patches, and writing directly to persistence memory or flash partitions, often requiring a reboot to apply changes. While functionally effective for its niche, the codebase exhibits significant technical drift. Compiled binaries are built with GCC 4.4.2 (circa 2010) and link against 2012-era OpenSSL and QNX libraries. Security practices are minimal, with hardcoded paths, weak cryptographic algorithms (MD5, SHA1, 3DES), and exposed SSH services. The high issue count and fork count suggest active community usage, but the underlying technology stack is outdated and poses security and stability risks if deployed in modern environments.