Adapting quickly to dynamic, uncertain environments—often called “open worlds” —remains a major challenge in robotics. Traditional Task and Motion Planning (TAMP) approaches struggle to cope with unforeseen changes, are data-inefficient when adapting, and do not leverage world models during learning. We address this issue with a hybrid planning and learning system that integrates two models: a low-level neural network-based model that learns stochastic transitions and drives exploration via an Intrinsic Curiosity Module (ICM), and a high-level symbolic planning model that captures abstract transitions using operators, enabling the agent to plan in an “imaginary” space and generate reward machines. Our evaluation in a robotic manipulation domain with sequential novelty injections demonstrates that our approach converges faster and outperforms state-of-the-art hybrid methods.

Open-world robotic tasks such as autonomous driving pose significant challenges to robot control due to unknown and unpredictable events that disrupt task performance. Neural network-based reinforcement learning (RL) techniques (like DQN, PPO, SAC, etc.) struggle to adapt in large domains and suffer from catastrophic forgetting. Hybrid planning and RL approaches have shown some promise in handling environmental changes but lack efficiency in accommodation speed. To address this limitation, we propose an enhanced hybrid system with a nested hierarchical action abstraction that can utilize previously acquired skills to effectively tackle unexpected novelties. We show that it can adapt faster and generalize better compared to state-of-the-art RL and hybrid approaches, significantly improving robustness when multiple environmental changes occur at the same time.

In dynamic open-world environments, agents continually face new challenges due to sudden and unpredictable novelties, hindering Task and Motion Planning (TAMP) in autonomous systems. We introduce a novel TAMP architecture that integrates symbolic planning with reinforcement learning to enable autonomous adaptation in such environments, operating without human guidance. Our approach employs symbolic goal representation within a goal-oriented learning framework, coupled with planner-guided goal identification, effectively managing abrupt changes where traditional reinforcement learning, re-planning, and hybrid methods fall short. Through sequential novelty injections in our experiments, we assess our method’s adaptability to continual learning scenarios. Extensive simulations conducted in a robotics domain corroborate the superiority of our approach, demonstrating faster convergence to higher performance compared to traditional methods. The success of our framework in navigating diverse novelty scenarios within a continuous domain underscores its potential for critical real-world applications.