🤖 AI Summary
This work addresses safe navigation of mobile robots in dynamic human-robot coexistence environments (e.g., homes, offices). We propose an online local trajectory planning method based on Model Predictive Control (MPC). Our key contribution is the first use of neural networks to estimate time-varying obstacle repulsive potential fields in real time, coupled with three dynamic obstacle modeling strategies—static snapshot, parallel prediction, and autoregressive prediction—to jointly optimize safety and computational efficiency. By integrating potential field methods with neural network-based modeling, our approach achieves superior performance over CIAO* and MPPI in the BenchMR simulator, satisfying stringent safety constraints while maintaining single-step planning latency under 100 ms. The method has been successfully deployed on a Husky UGV platform and validated in real-world dynamic office corridor scenarios, demonstrating robust and stable operation.
📝 Abstract
We address a task of local trajectory planning for the mobile robot in the presence of static and dynamic obstacles. Local trajectory is obtained as a numerical solution of the Model Predictive Control (MPC) problem. Collision avoidance may be provided by adding repulsive potential of the obstacles to the cost function of MPC. We develop an approach, where repulsive potential is estimated by the neural model. We propose and explore three possible strategies of handling dynamic obstacles. First, environment with dynamic obstacles is considered as a sequence of static environments. Second, the neural model predict a sequence of repulsive potential at once. Third, the neural model predict future repulsive potential step by step in autoregressive mode. We implement these strategies and compare it with CIAO* and MPPI using BenchMR framework. First two strategies showed higher performance than CIAO* and MPPI while preserving safety constraints. The third strategy was a bit slower, however it still satisfy time limits. We deploy our approach on Husky UGV mobile platform, which move through the office corridors under proposed MPC local trajectory planner. The code and trained models are available at url{https://github.com/CognitiveAISystems/Dynamic-Neural-Potential-Field}.