HanoiWorld : A Joint Embedding Predictive Architecture BasedWorld Model for Autonomous Vehicle Controller
This addresses the challenge of developing safer and more efficient autonomous vehicle controllers, though it appears incremental as it builds on existing JEPA approaches.
The study tackled the problem of data-demanding and unstable reinforcement learning for autonomous vehicle controllers by introducing HanoiWorld, a Joint Embedding Predictive Architecture (JEPA)-based world model using recurrent neural networks for long-term planning. Experiments on the Highway-Env package showed effective driving plans with safety-awareness and a considerable collision rate compared to state-of-the-art baselines.
Current attempts of Reinforcement Learning for Autonomous Controller are data-demanding while the results are under-performed, unstable, and unable to grasp and anchor on the concept of safety, and over-concentrating on noise features due to the nature of pixel reconstruction. While current Self-Supervised Learningapproachs that learning on high-dimensional representations by leveraging the JointEmbedding Predictive Architecture (JEPA) are interesting and an effective alternative, as the idea mimics the natural ability of the human brain in acquiring new skill usingimagination and minimal samples of observations. This study introduces Hanoi-World, a JEPA-based world model that using recurrent neural network (RNN) formaking longterm horizontal planning with effective inference time. Experimentsconducted on the Highway-Env package with difference enviroment showcase the effective capability of making a driving plan while safety-awareness, with considerablecollision rate in comparison with SOTA baselines