Agent faults pose a significant threat to the performance of multi-agent reinforcement learning (MARL) algorithms, introducing two key challenges. First, agents often struggle to extract critical information from the chaotic state space created by unexpected faults. Second, transitions recorded before and after faults in the replay buffer affect training unevenly, leading to a sample imbalance problem. To overcome these challenges, this paper enhances the fault tolerance of MARL by combining optimized model architecture with a tailored training data sampling strategy. Specifically, an attention mechanism is incorporated into the actor and critic networks to effectively and automatically detect fault information and dynamically regulate the attention given to faulty agents. Additionally, a prioritization mechanism is introduced to selectively sample transitions critical to current training needs. To further support research in this area, we design and open-source a highly decoupled code platform for fault-tolerant MARL, aimed at improving the efficiency of studying related problems. Experimental results demonstrate the effectiveness of our method in handling various types of faults, faults occurring in any agent, and faults arising at random times. Note to Practitioners—Multi-agent systems based on MARL outperform those using traditional control methods in terms of performance but remain highly vulnerable to unexpected faults. To improve fault tolerance in such systems, we introduce an attention mechanism that enables the neural network to dynamically adjust its focus on fault-related information. Additionally, a prioritization sampling strategy is employed to select critical samples from collected experiences that are most relevant to current training needs. Experimental results across various fault types demonstrate significant improvements in fault tolerance, validating the robustness of our approach. These findings suggest that the proposed method has the potential to be applied to real-world scenarios, such as multi-robot systems and autonomous vehicle fleets.

Aiming to address the core issue that the current intelligence evaluation for Unmanned Ground Vehicles (UGVs) overly rely on static performance metrics and lack dynamic quantitative characterization of learning internalization capability (LIC), this study proposes a dynamic evaluation framework based on time series analysis. The framework begins by constructing a multidimensional test scenario parameter system and collecting externally observable performance sequence data. It then introduces a sliding window-based slope-standard deviation collaborative analysis technique to achieve unsupervised division of learning phases, from which five core evaluation metrics are extracted to comprehensively quantify the multidimensional dynamic characteristics of LIC in terms of efficiency, stability, and overall effectiveness. Simulation experiments were carried out using UGVs equipped with three types of path-planning algorithms in low-, medium-, and high-difficulty scenarios. Results demonstrate that the proposed algorithm can effectively distinguish multi-dimensional differences in LIC among different UGVs, exhibiting strong discriminative power and interpretability. This study provides a standardized evaluation tool for UGV intelligent selection, algorithm iteration optimization, and training strategy design, and offering significant reference value for the evaluation of the learnability of autonomous driving systems.

Recent advances in decision-making policies have led to notable progress in various fields such as autonomous driving and robotics. However, testing these policies remains crucial with the existence of critical scenarios that may threaten their reliability. Despite ongoing research, challenges such as low testing efficiency and limited scenario diversity persist, primarily due to the complexity of the policies and their operating environments. To address these challenges, this paper proposes an adaptable Large Language Model (LLM)-driven adaptive testing framework to explore critical and diverse testing scenarios for decision-making policies. Specifically, we introduce a “generate-test-feedback” pipeline, leveraging templated prompt engineering to harness the world knowledge and reasoning abilities of LLMs. Additionally, a multi-scale scenario generation strategy is proposed to overcome the limitations of LLMs in making fine-grained adjustments, thereby further enhancing testing efficiency. Furthermore, during the feedback phase, a well-designed scenario evaluation module assesses the criticality and diversity of the generated scenarios for adaptive testing. The proposed LLM-driven method is extensively evaluated on five widely recognized benchmarks, demonstrating that our proposed method significantly outperforms baseline methods in uncovering both critical and diverse scenarios. These findings highlight the great promise of LLM-driven schemes in advancing the testing of decision-making policies.

Multi-Agent Reinforcement Learning (MARL) holds significant promise for enabling cooperative driving among Connected and Automated Vehicles (CAVs). However, its practical application is hindered by a critical limitation, i.e., insufficient fault tolerance against observational faults. Such faults, which appear as perturbations in the vehicles' perceived data, can substantially compromise the performance of MARL-based driving systems. Addressing this problem presents two primary challenges. One is to generate adversarial perturbations that effectively stress the policy during training, and the other is to equip vehicles with the capability to mitigate the impact of corrupted observations. To overcome the challenges, we propose a fault-tolerant MARL method for cooperative on-ramp vehicles incorporating two key agents. First, an adversarial fault injection agent is co-trained to generate perturbations that actively challenge and harden the vehicle policies. Second, we design a novel fault-tolerant vehicle agent equipped with a self-diagnosis capability, which leverages the inherent spatio-temporal correlations in vehicle state sequences to detect faults and reconstruct credible observations, thereby shielding the policy from misleading inputs. Experiments in a simulated highway merging scenario demonstrate that our method significantly outperforms baseline MARL approaches, achieving near-fault-free levels of safety and efficiency under various observation fault patterns.

Testing and evaluating decision-making agents remains challenging due to unknown system architectures, limited access to internal states, and the vastness of high-dimensional scenario spaces. Existing testing approaches often rely on surrogate models of decision-making agents to generate large-scale scenario libraries; however, discrepancies between surrogate models and real decision-making agents significantly limit their generalizability and practical applicability. To address this challenge, this paper proposes intelligent resilience testing (IRTest), a unified online adaptive testing framework designed to rapidly adjust to diverse decision-making agents. IRTest initializes with an offline-trained surrogate prediction model and progressively reduces surrogate-to-real gap during testing through two complementary adaptation mechanisms: (i) online neural fine-tuning in data-rich regimes, and (ii) lightweight importance-sampling-based weighting correction in data-limited regimes. A Bayesian optimization strategy, equipped with bias-corrected acquisition functions, guides scenario generation to balance exploration and exploitation in complex testing spaces. Extensive experiments across varying levels of task complexity and system heterogeneity demonstrate that IRTest consistently improves failure-discovery efficiency, testing robustness, and cross-system generalizability. These results highlight the potential of IRTest as a practical solution for scalable, adaptive, and resilient testing of decision-making agents.