Automating scientific discovery in complex, experiment-driven domains requires more than iterative mutation of programs; it demands structured hypothesis management, environment interaction, and principled reflection. We present OR-Agent, a configurable multi-agent research framework designed for automated exploration in rich experimental environments. OR-Agent organizes research as a structured tree-based workflow that explicitly models branching hypothesis generation and systematic backtracking, enabling controlled management of research trajectories beyond simple mutation-crossover loops. At its core, we introduce an evolutionary-systematic ideation mechanism that unifies evolutionary selection of research starting points, comprehensive research plan generation, and coordinated exploration within a research tree. We introduce a hierarchical optimization-inspired reflection system in which short-term reflections act as verbal gradients, long-term reflections as verbal momentum, and memory compression as semantic weight decay, collectively forming a principled mechanism for governing research dynamics. We conduct extensive experiments across classical combinatorial optimization benchmarks as well as simulation-based cooperative driving scenarios. Results demonstrate that OR-Agent outperforms strong evolutionary baselines while providing a general, extensible, and inspectable framework for AI-assisted scientific discovery. All code and experimental data are publicly available at https://github.com/qiliuchn/OR-Agent.

翻译：在复杂、实验驱动的领域中实现科学发现的自动化，需要的不仅仅是程序的迭代变异；它要求结构化的假设管理、环境交互以及原则性的反思。我们提出了OR-Agent，一个可配置的多智能体研究框架，专为在丰富的实验环境中进行自动化探索而设计。OR-Agent将研究组织为一个结构化的基于树的工作流，显式地建模分支假设生成和系统性回溯，从而实现对研究轨迹的受控管理，超越了简单的变异-交叉循环。其核心是，我们引入了一种进化-系统性构思机制，该机制将研究起点的进化选择、全面研究计划的生成以及研究树内的协同探索统一起来。我们引入了一个受分层优化启发的反思系统，其中短期反思充当言语梯度，长期反思充当言语动量，记忆压缩充当语义权重衰减，共同构成了一个管理研究动态的原则性机制。我们在经典组合优化基准以及基于模拟的协同驾驶场景中进行了广泛的实验。结果表明，OR-Agent优于强大的进化基线，同时为AI辅助的科学发现提供了一个通用、可扩展且可检查的框架。所有代码和实验数据均在 https://github.com/qiliuchn/OR-Agent 公开提供。