Petter Ogren | Searcharxiv

arXiv2026

Marinarium: A Modular Experimental Facility for Reproducible Maritime and Space-Analog Field Robotics

Field robotics research in maritime and space domains is constrained by a persistent gap between low-cost, low-fidelity simulation and costly offshore experimentation. Instrumented water tanks partially bridge this gap but often provide limited sensing, restricted experimental capabilities, and weak integration with simulation tools. To address these limitations, we present Marinarium, a modular, standalone experimental facility that provides a cost-effective intermediate testbed between simulation and field deployment. Marinarium combines a fully instrumented underwater and aerial operational volume with motion capture (MoCap), a retractable roof enabling both sheltered and open-air operation, a digital twin implemented in SMaRCSim, and direct integration with a planar space robotics laboratory, enabling both maritime and underwater space-analog experimentation. We present the design rationale of the facility and validate its capabilities through four representative studies in field robotics: (i) data-driven system identification of underwater vehicle dynamics; (ii) heterogeneous multi-domain robotic rendezvous; (iii) sim-to-real transfer for underwater robotics using learned dynamics residuals; and (iv) cross-domain validation of spacecraft autonomy using underwater surrogates. Together, these studies demonstrate that Marinarium enables reproducible, instrumented experimentation across multiple field robotics challenges that would otherwise require costly offshore deployments or be impractical to investigate using simulation alone.

cs.RO↗

arXiv2025

Learned Controllers for Agile Quadrotors in Pursuit-Evasion Games

In this letter we study 1v1 quadrotor pursuit-evasion, where a pursuer and an evader are trained via reinforcement learning (RL) by competing against each other. Such adversarial settings face well-known challenges: each agent's policy changes during training, creating a non-stationary environment; agents might overfit to the current opponent and forget earlier strategies (catastrophic forgetting); and the competitive dynamics can cause strategy cycling or policy collapse. To address these issues, we propose Asynchronous Multi-Stage Population-Based training with Hedge sampling (AMSPBH), a method based on Policy-Space Response Oracles (PSRO) and adapted to quadrotor RL control. PSRO maintains a population of previously trained policies and trains new approximate best responses against mixtures of that population instead of against a single opponent. In AMSPBH, each generation trains one agent with Proximal Policy Optimization (PPO) against frozen opponent policies, while a Hedge sampler assigns higher probability to opponents that are currently difficult to beat. We show that: (i) AMSPBH discovers new strategies while retaining competence against older opponents, reaching a regime where additional best-response training gives limited improvement; (ii) compared to training against only the latest opponent, population-based training generalizes better across diverse and unseen strategies; and (iii) the learned population contains distinct pursuit and evasion behaviors, providing useful strategic diversity for finding weaknesses and improving controller robustness. We validate the trained policies with hardware experiments on Crazyflie brushless quadrotors, showing zero-shot sim-to-real transfer of agile, reactive pursuit-evasion behavior against both handcrafted and learned adversaries, with physical flights reaching up to 4.87 m/s.

cs.RO↗