# Dynamic Neural Potential Field: Online Trajectory Optimization in the Presence of Moving Obstacles

arXiv: https://arxiv.org/abs/2410.06819 · HTML: https://mlsanity.com/paper/2410.06819/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 23 of its 59 references that are in the map.
- First posted on arXiv 9 Oct 2024.

## Builds on (most cited first, 23 of 23)

- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Habitat 2.0: Training Home Assistants to Rearrange their Habitat](https://mlsanity.com/paper/2106.14405/) (2021)
- [Vision-Only Robot Navigation in a Neural Radiance World](https://mlsanity.com/paper/2110.00168/) (2021)
- [acados: a modular open-source framework for fast embedded optimal control](https://mlsanity.com/paper/1910.13753/) (2019)
- [RTAB-Map as an Open-Source Lidar and Visual SLAM Library for Large-Scale and Long-Term Online Operation](https://mlsanity.com/paper/2403.06341/) (2024)
- [Safety-Critical Model Predictive Control with Discrete-Time Control Barrier Function](https://mlsanity.com/paper/2007.11718/) (2020)
- [Real-time Neural-MPC: Deep Learning Model Predictive Control for Quadrotors and Agile Robotic Platforms](https://mlsanity.com/paper/2203.07747/) (2022)
- [Discrete Residual Flow for Probabilistic Pedestrian Behavior Prediction](https://mlsanity.com/paper/1910.08041/) (2019)
- [Theta*: Any-Angle Path Planning on Grids](https://mlsanity.com/paper/1401.3843/) (2014)
- [Learning for CasADi: Data-driven Models in Numerical Optimization](https://mlsanity.com/paper/2312.05873/) (2023)
- [Autonomous Navigation of AGVs in Unknown Cluttered Environments: log-MPPI Control Strategy](https://mlsanity.com/paper/2203.16599/) (2022)
- [OFMPNet: Deep End-to-End Model for Occupancy and Flow Prediction in Urban Environment](https://mlsanity.com/paper/2404.02263/) (2024)
- [Safety-Critical Control and Planning for Obstacle Avoidance between Polytopes with Control Barrier Functions](https://mlsanity.com/paper/2109.12313/) (2021)
- [N-MPC for Deep Neural Network-Based Collision Avoidance exploiting Depth Images](https://mlsanity.com/paper/2402.13038/) (2024)
- [Safe Interval Path Planning With Kinodynamic Constraints](https://mlsanity.com/paper/2302.00776/) (2023)
- [Neural Potential Field for Obstacle-Aware Local Motion Planning](https://mlsanity.com/paper/2310.16362/) (2023)
- [An NMPC Approach using Convex Inner Approximations for Online Motion Planning with Guaranteed Collision Avoidance](https://mlsanity.com/paper/1909.08267/) (2019)
- [CIAO^star: MPC-based Safe Motion Planning in Predictable Dynamic Environments](https://mlsanity.com/paper/2001.05449/) (2020)
- [Model Predictive Path Integral Control Framework for Partially Observable Navigation: A Quadrotor Case Study](https://mlsanity.com/paper/2004.08641/) (2020)
- [Search-Based Online Trajectory Planning for Car-like Robots in Highly Dynamic Environments](https://mlsanity.com/paper/2011.03664/) (2020)
- [Non-Holonomic RRT & MPC: Path and Trajectory Planning for an Autonomous Cycle Rickshaw](https://mlsanity.com/paper/2103.06141/) (2021)
- [Path Integral Control with Rollout Clustering and Dynamic Obstacles](https://mlsanity.com/paper/2403.18066/) (2024)
- [DNFOMP: Dynamic Neural Field Optimal Motion Planner for Navigation of Autonomous Robots in Cluttered Environment](https://mlsanity.com/paper/2308.03539/) (2023)

## Abstract (arXiv)

Generalist robot policies must operate safely and reliably in everyday human environments such as homes, offices, and warehouses, where people and objects move unpredictably. We present Dynamic Neural Potential Field (NPField-GPT), a learning-enhanced model predictive control (MPC) framework that couples classical optimization with a Transformer-based predictor of footprint-aware repulsive potentials. Given an occupancy sub-map, robot footprint, and optional dynamic-obstacle cues, our NPField-GPT model forecasts a horizon of differentiable potentials that are injected into a sequential quadratic MPC program via L4CasADi, yielding real-time, constraint-aware trajectory optimization. We additionally study two baselines: NPField-StaticMLP, where a dynamic scene is treated as a sequence of static maps; and NPField-DynamicMLP, which predicts the future potential sequence in parallel with an MLP. In dynamic indoor scenarios from BenchMR and on a Husky UGV in office corridors, NPField-GPT produces more efficient and safer trajectories under motion changes, while StaticMLP/DynamicMLP offer lower latency. We also compare with the CIAO* and MPPI baselines. Across methods, the Transformer+MPC synergy preserves the transparency and stability of model-based planning while learning only the part that benefits from data: spatiotemporal collision risk. Code and trained models are available at https://github.com/CognitiveAISystems/Dynamic-Neural-Potential-Field