Overview of Improvement Methods for Three Evaluation Indicators of Path Planning A * Algorithm

Abstract

In the era of rapid development of industrial automation in the 21st century, robots are currently the focus of global technological development. The efficiency and safety of robot work are indicators that people evaluate their work, and most robots need to move during work, so it is necessary to study the problem of robot movement. The key to robot movement problem is route planning, and the research topic of this article is three improvement methods according to the conventional A * algorithm. This article first introduces the conventional A * algorithm, and then analyzes three evaluation indicators of path search time, amount of turning points, and turning flatness of the traditional A * algorithm. It is found that these indicators are the main factors affecting the efficiency of car search. Thus, several typical improvement methods were demonstrated, such as dynamically adjusting the cost estimation function method, dynamically adjusting the search domain method, and geometric optimization method. After mathematical modeling and formula analysis of these three methods, this article concludes that the traditional A * algorithm is not suitable for working in actual working conditions with randomly distributed obstacles. After improving the above methods, the improved A * algorithm has significant implications for path search efficiency and safety.