With the rapid development and the growing deployment of autonomous ground
vehicles (AGVs) worldwide, there is an increasing need to design reliable,
efficient, robust, and scalable motion-planning algorithms. These algorithms are
crucial for fulfilling the desired goals of safety, comfort, efficiency, and
accessibility. To design optimal motion-planning algorithms, it is beneficial to
explore existing techniques and make improvements by addressing the limitations
of associated techniques, utilizing hybrid algorithms, or developing novel
strategies. This article categorizes and overviews numerous motion-planning
algorithms for AGVs, shedding light on their strengths and weaknesses for a
comprehensive understanding. For various applications of AGVs, such as urban and
off-road autonomous driving, the features of driving conditions and vehicle
kinodynamics are outlined, and sample-tailored motion-planning algorithms built
upon relevant canonical techniques are briefly introduced. As a result of the
overview, future research efforts on motion-planning techniques are identified
and discussed.
