Implementation of the A* Algorithm for Determining the Best Route for an Autonomous Electric Vehicle

In undertaking a journey to reach a destination, every vehicle, including autonomous ones, requires a route. However, autonomous vehicles face challenges in finding the best route, especially for routes that may be obstructed due to road damage, as seen in Indonesia. Thus, this study aims to determine the optimal route by selecting the most efficient travel path using the A* algorithm. The data is obtained from a GPS sensor placed in the autonomous vehicle. Simulation tests are conducted within the Universitas Sriwijaya Campus in Palembang, involving 19 nodes, and the result is a distance of 179.28198978430741 meters from node 0 to node 3. Real-time testing is carried out within the Universitas Sriwijaya Campus in Indralaya, involving 42 nodes, and the resulting travel distance is 346.792308027204274 meters from node 0 to node 5. In the simulation testing, an average error of 3.9429109681873543 meters is observed, while in real-time testing, the error averages 4.1706054421860576 meters. Further testing is conducted by introducing obstacles on the road, prompting the system to perform rerouting. Originally, the autonomous electric vehicle’s route passes through nodes 5 to 6 to 7 to 27 to 28 but is altered to 5 to 6 to 9 to 27 to 28. Based on the conducted tests, the A* algorithm proves effective in finding the best routes for autonomous electric vehicles, both in normal conditions and when encountering obstacles.