Project Case Study

TeamX – Intent & Trajectory Prediction

An AI-powered trajectory prediction system developed for the MAHE Mobility Challenge 2025 to enhance the safety of autonomous vehicles in complex urban environments. Using a Spatio-Temporal Graph Convolutional Neural Network (ST-GCNN), the platform predicts pedestrian and cyclist movements up to three seconds into the future by modeling social interactions, crowd behavior, and motion dynamics. The solution leverages real-world autonomous driving data from the nuScenes dataset to help self-driving vehicles anticipate intent, avoid collisions, and make safer navigation decisions in real time.