Path Planning Acceleration with GPU for an Omnidirectional Mobile Robot

Resumen

Path planning in a state space using the iterative deepening method is a complex problem that can be accelerated using a GPU. In this approach, the state space is divided into smaller subspaces and iterative depth search is applied to each of these. The parallel capabilities of the GPU are utilized to process several subspaces concurrently. Furthermore, the shared memory in the GPU can be leveraged to store relevant data and reduce access time to the global memory. Implementing this approach on the GPU can provide significant acceleration compared to CPU execution. However, careful optimization and parameter tuning are required to utilize the GPU’s capacity fully. In addition to a detailed description of the proposed methodology, experimental results are presented that demonstrate the superiority of our approach compared to traditional CPU-based methods. These results highlight the potential of GPUs to transform trajectory planning in mobile robots, offering a route to faster and more efficient solutions. Trajectory planning in state spaces represents a significant challenge in mobile robotics, particularly in applications that demand fast and efficient responses in dynamic and complex environments. This work introduces a novel method to accelerate route planning in an omnidirectional mobile robot fusing advances in hardware with sophisticated algorithmic techniques, a new paradigm is established in path planning for omnidirectional mobile robots, marking an important milestone in the search for more agile and capable robotic systems.