The project aims to develop a navigation platform for autonomous vehicles through an obstacle course, based on a physical model of vehicle motion. The solution integrates numerical methods for solving motion equations or Neural Networks to generate optimal paths in terms of distance and time. The platform considers vehicle dynamics based on Ackerman Steering and allows for real-time navigation solutions.
The project employs the PRM (Probabilistic Roadmap) algorithm to generate a sparse graph of valid configurations in the state space (x, y, θ) and connect them with edges that represent the vehicle’s dynamics, calculated using the Ackermann steering model. The solution is based on different models, including a numerical approach and several neural network-based models, and is evaluated in terms of path quality, computational time, and optimal distance.
Additionally, a comparison is made between these methods to determine which offers the best balance between real-time performance and solution accuracy. The final solution includes a platform that provides visualization and feedback on runtime and path quality, making it suitable for real-world autonomous vehicle navigation scenarios.