Search for: All records

Robot teleoperation with extended reality (XR teleoperation) enables intuitive interaction by mapping user motions to remote robots with real-time 3D feedback. However, existing systems suffer from large completion delays and trajectory deviations under prolonged network latency, rooted in their exclusive reliance on network communication and strict synchronous execution architecture. Moreover, network fluctuations destabilize teleoperation accuracy, while dynamic user motions amplify teleoperation errors. We present MATER, an end-to-end XR teleoperation framework that introduces a mutually-aware architecture in which each side reconstructs its counterpart's delayed or missing state to decouple the execution from network dependency. MATER includes latency-adaptive input window and user motion gap interpolation techniques to handle unstable network communication. It also proposes motion-driven robot state rollback and robot trajectory coordination to handle complex motions. The key idea behind these techniques is to adapt on the fly by reshaping reconstruction and filling gaps as network conditions fluctuate, and by realigning states when motions become fast or complex. Together with lightweight local synchronization and bandwidth optimizations, these system-level advances make MATER resilient to both network and motion dynamics. We implement MATER across three hardware settings, including simulated and physical robots, and evaluate it on 9,500 real-world teleoperation trials from the RoboSet dataset [1], covering single- and multi-step missions. Compared to state-of-the-art XR teleoperation frameworks, MATER reduces teleoperation error by up to 69.8% on WLAN and 73.1% on cellular networks with only 6.7% maximum runtime overhead. It also shortens mission completion time by up to 47.7%, enabling smoother teleoperation. A real-world case study on ten stationary and mobile missions further shows MATER achieves up to 37.7% faster completion while lowering average teleoperation error by up to 57.2%. MATER code is available at: https://github.com/rtenlab/mater 

Agentic AI aims to create systems that set their own goals, adapt proactively to change, and refine behavior through continuous experience. Recent advances suggest that, when facing multiple and unforeseen tasks, agents could benefit from sharing machine-learned knowledge and reusing policies that have already been fully or partially learned by other agents. However, how to query, select, and retrieve policies from a pool of agents, and how to integrate such policies remains a largely unexplored area. This study explores how an agent decides what knowledge to select, from whom, and when and how to integrate it in its own policy in order to accelerate its own learning. The proposed algorithm, Modular Sharing and Composition in Collective Learning (MOSAIC), improves learning in agentic collectives by combining (1) knowledge selection using performance signals and cosine similarity on Wasserstein task embeddings, (2) modular and transferable neural representations via masks, and (3) policy integration, composition and fine-tuning. MOSAIC outperforms isolated learners and global sharing approaches in both learning speed and overall performance, and in some cases solves tasks that isolated agents cannot. The results also demonstrate that selective, goal-driven reuse leads to less susceptibility to task interference. We also observe the emergence of self-organization, where agents solving simpler tasks accelerate the learning of harder ones through shared knowledge. 

Abstract Multispecies mutualistic interactions are ubiquitous and essential in nature, yet they face several threats, many of which have been exacerbated in the Anthropocene era. Understanding the factors that drive the stability and persistence of mutualism has become increasingly important in light of global change. Although dispersal is widely recognized as a crucial spatially explicit process in maintaining biodiversity and community structure, knowledge about how the dispersal of mutualists contributes to the persistence of mutualistic systems remains limited. In this study, we used a synthetic mutualism formed by genetically modified budding yeast to investigate the effect of dispersal on the persistence and stability of mutualisms under exploitation. We found that dispersal increased the persistence of exploited mutualisms by 80% compared to the isolated systems. Furthermore, our results showed that dispersal increased local diversity, decreased beta diversity among local communities, and stabilized community structure at the regional scale. Our results indicate that dispersal can allow mutualisms to persist in meta-communities by reintroducing species that are locally competitively excluded by exploiters. With limited dispersal, e.g. due to increased fragmentation of meta-communities, mutualisms might be more prone to breakdown. Taken together, our results highlight the critical role of dispersal in facilitating the persistence of mutualism. 

Coevolution is a ubiquitous driver of diversification in both mutualistic and antagonistic interactions between species. In mutualisms, coevolution can result in trait complementarity between partners that facilitates their persistence. Despite its importance, most of what we know about coevolution in mutualism comes from obligate interactions, whereas we know comparatively little about facultative interactions, arguably the most common type of mutualism. To evaluate coevolutionary dynamics in facultative mutualism and test how it compares with obligate mutualisms, we used a synthetic yeast mutualism where the partners exchange essential nutrient resources. We manipulated mutualism dependency by controlling the availability of mutualistic resources in the environment and measured coevolution via time-shift assays and tracking the evolution of mutualistic traits over time. In addition, we genotyped the evolved and ancestral mutualists to test for differences in the strength of coevolutionary selection between facultative and obligate mutualisms. We found evidence of coevolution in both facultative and obligate mutualisms, but coevolution was weaker and slower in facultative mutualisms. We also found evidence for evolution of trait complementarity in obligate mutualisms but not in facultative mutualisms. Furthermore, obligate mutualists had more SNPs under positive selection than facultative mutualists. Together, these results provide strong evidence that mutualism dependency impacts both the strength of coevolution and the rate of trait evolution.