Key Ingredients of Self-Driving Cars

TL;DR

The paper presents a concise,; comprehensive overview of the key ingredients of autonomous cars, spanning driving automation levels, sensors, software modules, open-source datasets, industry leaders, applications, and challenges. It surveys architectural foundations and typical subsystem roles—perception, localization and mapping, prediction, planning, and control—along with representative datasets (e.g., Cityscapes, KITTI, ApolloScape) and industry milestones (Waymo, Tesla, GM). It highlights model-based and data-driven approaches across perception, SLAM, prediction, and planning, and discusses practical control strategies (PID, LQR, MPC) and their hybrids. The discussion identifies critical bottlenecks in perception robustness, long-term SLAM stability, real-time fusion of heterogeneous sensors, and socio-ethical acceptance, framing an integrated view across subsystems as essential for advancing deployment and safety in autonomous driving.

Abstract

Over the past decade, many research articles have been published in the area of autonomous driving. However, most of them focus only on a specific technological area, such as visual environment perception, vehicle control, etc. Furthermore, due to fast advances in the self-driving car technology, such articles become obsolete very fast. In this paper, we give a brief but comprehensive overview on key ingredients of autonomous cars (ACs), including driving automation levels, AC sensors, AC software, open source datasets, industry leaders, AC applications and existing challenges.

Figures (2)

• Figure 1: Numbers of publications and citations in autonomous driving research over the past decade.
• Figure 2: Software architecture of our proposed ADS.