For efficient and safe autonomous driving, it is essential that autonomous vehicles can predict the motion of other traffic agents. While highly accurate, current motion prediction models often impose significant challenges in terms of training resource requirements and deployment on embedded hardware. We propose a new efficient motion prediction model, which achieves highly competitive benchmark results while training only a few hours on a single GPU. Due to our lightweight architectural choices and the focus on reducing the required training resources, our model can easily be applied to custom datasets. Furthermore, its low inference latency makes it particularly suitable for deployment in autonomous applications with limited computing resources.

Multi-UAV pursuit-evasion, where pursuers aim to capture evaders, poses a key challenge for UAV swarm intelligence. Multi-agent reinforcement learning (MARL) has demonstrated potential in modeling cooperative behaviors, but most RL-based approaches remain constrained to simplified simulations with limited dynamics or fixed scenarios. Previous attempts to deploy RL policy to real-world pursuit-evasion are largely restricted to two-dimensional scenarios, such as ground vehicles or UAVs at fixed altitudes. In this paper, we address multi-UAV pursuit-evasion by considering UAV dynamics and physical constraints. We introduce an evader prediction-enhanced network to tackle partial observability in cooperative strategy learning. Additionally, we propose an adaptive environment generator within MARL training, enabling higher exploration efficiency and better policy generalization across diverse scenarios. Simulations show our method significantly outperforms all baselines in challenging scenarios, generalizing to unseen scenarios with a 100% capture rate. Finally, we derive a feasible policy via a two-stage reward refinement and deploy the policy on real quadrotors in a zero-shot manner. To our knowledge, this is the first work to derive and deploy an RL-based policy using collective thrust and body rates control commands for multi-UAV pursuit-evasion in unknown environments. The open-source code and videos are available at https://sites.google.com/view/pursuit-evasion-rl.

In the field of locomotion task of quadruped robots, Blind Policy and Perceptive Policy each have their own advantages and limitations. The Blind Policy relies on preset sensor information and algorithms, suitable for known and structured environments, but it lacks adaptability in complex or unknown environments. The Perceptive Policy uses visual sensors to obtain detailed environmental information, allowing it to adapt to complex terrains, but its effectiveness is limited under occluded conditions, especially when perception fails. Unlike the Blind Policy, the Perceptive Policy is not as robust under these conditions. To address these challenges, we propose a MBC:Multi-Brain collaborative system that incorporates the concepts of Multi-Agent Reinforcement Learning and introduces collaboration between the Blind Policy and the Perceptive Policy. By applying this multi-policy collaborative model to a quadruped robot, the robot can maintain stable locomotion even when the perceptual system is impaired or observational data is incomplete. Our simulations and real-world experiments demonstrate that this system significantly improves the robot's passability and robustness against perception failures in complex environments, validating the effectiveness of multi-policy collaboration in enhancing robotic motion performance.

Multi-human multi-robot teams combine the complementary strengths of humans and robots to tackle complex tasks across diverse applications. However, the inherent heterogeneity of these teams presents significant challenges in initial task allocation (ITA), which involves assigning the most suitable tasks to each team member based on their individual capabilities before task execution. While current learning-based methods have shown promising results, they are often computationally expensive to train, and lack the flexibility to incorporate user preferences in multi-objective optimization and adapt to last-minute changes in real-world dynamic environments. To address these issues, we propose REBEL, an LLM-based ITA framework that integrates rule-based and experience-enhanced learning. By leveraging Retrieval-Augmented Generation, REBEL dynamically retrieves relevant rules and past experiences, enhancing reasoning efficiency. Additionally, REBEL can complement pre-trained RL-based ITA policies, improving situational awareness and overall team performance. Extensive experiments validate the effectiveness of our approach across various settings. More details are available at https://sites.google.com/view/ita-rebel .

Although language model (LM) agents are demonstrating growing potential in many domains, their success in cybersecurity has been limited due to simplistic design and the lack of fundamental features for this domain. We present EnIGMA, an LM agent for autonomously solving Capture The Flag (CTF) challenges. EnIGMA introduces new Agent-Computer Interfaces (ACIs) to improve the success rate on CTF challenges. We establish the novel Interactive Agent Tool concept, which enables LM agents to run interactive command-line utilities essential for these challenges. Empirical analysis of EnIGMA on over 350 CTF challenges from three different benchmarks indicates that providing a robust set of new tools with demonstration of their usage helps the LM solve complex problems and achieves state-of-the-art results on the NYU CTF and Intercode-CTF benchmarks. Finally, we discuss insights on ACI design and agent behavior on cybersecurity tasks that highlight the need to adapt real-world tools for LM agents.

Developing AI agents powered by large language models (LLMs) faces significant challenges in achieving true Turing completeness and adaptive, code-driven evolution. Current approaches often generate code independently of its runtime context, relying heavily on the LLM's memory, which results in inefficiencies and limits adaptability. Manual protocol development in sandbox environments further constrains the agent's autonomous adaptability. Crucially, achieving consistency in code and context across multi-turn interactions and ensuring isolation of local variables within each interaction remains an unsolved problem. We introduce MOSS (llM-oriented Operating System Simulation), a novel framework that addresses these challenges by integrating code generation with a dynamic context management system. MOSS ensures consistency and adaptability by using a mechanism that maintains the Python context across interactions, including isolation of local variables and preservation of runtime integrity. At its core, the framework employs an Inversion of Control (IoC) container in conjunction with decorators to enforce the least knowledge principle, allowing agents to focus on abstract interfaces rather than concrete implementations. This facilitates seamless integration of new tools and libraries, enables runtime instance replacement, and reduces prompt complexity, providing a "what you see is what you get" environment for the agent. Through a series of case studies, we show how this framework can enhance the efficiency and capabilities of agent development and highlight its advantages in moving towards Turing-complete agents capable of evolving through code.

Trust is essential in human-robot collaboration. Even more so in multi-human multi-robot teams where trust is vital to ensure teaming cohesion in complex operational environments. Yet, at the moment, trust is rarely considered a factor during task allocation and reallocation in algorithms used in multi-human, multi-robot collaboration contexts. Prior work on trust in single-human-robot interaction has identified that including trust as a parameter in human-robot interaction significantly improves both performance outcomes and human experience with robotic systems. However, very little research has explored the impact of trust in multi-human multi-robot collaboration, specifically in the context of task allocation. In this paper, we introduce a new trust model, the Expectation Comparison Trust (ECT) model, and employ it with three trust models from prior work and a baseline no-trust model to investigate the impact of trust on task allocation outcomes in multi-human multi-robot collaboration. Our experiment involved different team configurations, including 2 humans, 2 robots, 5 humans, 5 robots, and 10 humans, 10 robots. Results showed that using trust-based models generally led to better task allocation outcomes in larger teams (10 humans and 10 robots) than in smaller teams. We discuss the implications of our findings and provide recommendations for future work on integrating trust as a variable for task allocation in multi-human, multi-robot collaboration.

Large crowds exhibit intricate behaviors and significant emergent properties, yet existing crowd simulation systems often lack behavioral diversity, resulting in homogeneous simulation outcomes. To address this limitation, we propose incorporating anisotropic fields (AFs) as a fundamental structure for depicting the uncertainty in crowd movement. By leveraging AFs, our method can rapidly generate crowd simulations with intricate behavioral patterns that better reflect the inherent complexity of real crowds. The AFs are generated either through intuitive sketching or extracted from real crowd videos, enabling flexible and efficient crowd simulation systems. We demonstrate the effectiveness of our approach through several representative scenarios, showcasing a significant improvement in behavioral diversity compared to classical methods. Our findings indicate that by incorporating AFs, crowd simulation systems can achieve a much higher similarity to real-world crowd systems. Our code is publicly available at https://github.com/tomblack2014/AF\_Generation.

Accurately predicting the trajectory of surrounding vehicles is a critical challenge for autonomous vehicles. In complex traffic scenarios, there are two significant issues with the current autonomous driving system: the cognitive uncertainty of prediction and the lack of risk awareness, which limit the further development of autonomous driving. To address this challenge, we introduce a novel trajectory prediction model that incorporates insights and principles from driving behavior, ethical decision-making, and risk assessment. Based on joint prediction, our model consists of interaction, intention, and risk assessment modules. The dynamic variation of interaction between vehicles can be comprehensively captured at each timestamp in the interaction module. Based on interaction information, our model considers primary intentions for vehicles to enhance the diversity of trajectory generation. The optimization of predicted trajectories follows the advanced risk-aware decision-making principles. Experimental results are evaluated on the DeepAccident dataset; our approach shows its remarkable prediction performance on normal and accident scenarios and outperforms the state-of-the-art algorithms by at least 28.9\% and 26.5\%, respectively. The proposed model improves the proficiency and adaptability of trajectory prediction in complex traffic scenarios. The code for the proposed model is available at https://sites.google.com/view/ir-prediction.

Perception still remains a challenging problem for autonomous navigation in unknown environment, especially for aerial vehicles. Most mapping algorithms for autonomous navigation are specifically designed for their very intended task, which hinders extended usage or cooperative task. In this paper, we propose a voxel mapping system that can build an adaptable map for multiple tasks. The system employs hash table-based map structure and manages each voxel with spatial and temporal priorities without explicit map boundary. We also introduce an efficient map-sharing feature with minimal bandwidth to enable multi-agent applications. We tested the system in real world and simulation environment by applying it for various tasks including local mapping, global mapping, cooperative multi-agent navigation, and high-speed navigation. Our system proved its capability to build customizable map with high resolution, wide coverage, and real-time performance regardless of sensor and environment. The system can build a full-resolution map using the map-sharing feature, with over 95 % of bandwidth reduction from raw sensor data.