Approaching Current Challenges in Developing a Software Stack for Fully Autonomous Driving

Personal use of this material is permitted. Abstract -- Autonomous driving is a complex undertaking. A common approach is to break down the driving task into individual subtasks through modularization. These sub-modules are usually developed and published separately. However, if these individually developed algorithms have to be combined again to form a full-stack autonomous driving software, this poses particular challenges. Drawing upon our practical experience in developing the software of TUM Autonomous Motorsport, we have identified and derived these challenges in developing an autonomous driving software stack within a scientific environment. We do not focus on the specific challenges of individual algorithms but on the general difficulties that arise when deploying research algorithms on real-world test vehicles. T o overcome these challenges, we introduce strategies that have been effective in our development approach. We additionally provide open-source implementations that enable these concepts on GitHub. As a result, this paper's contributions will simplify future full-stack autonomous driving projects, which are essential for a thorough evaluation of the individual algorithms. Autonomous driving is a rapidly growing research field with a continuously increasing number of publications [1]. However, only a few algorithms and approaches have been deployed in real-world applications. A recent review paper [2] shows that out of 111 publications in the area of motion planning and vehicle control, only 44 % have been deployed in an autonomous driving software stack.