Search for: All records

In this paper, we introduce a model of evolution and learning in robots that co-optimizes a distribution of latent design vectors (genotypes) and a mixture of control experts (neural modules), which are gated by the latent coordinates of each decoded design (phenotype). This provides a scalable alternative to co-design algorithms that either train an individual policy for every robot, which is inefficient, or a monolithic universal controller for all robots, which results in overly conservative structures and behaviors. Our approach lies somewhere between these two extremes, preserving ancestral knowledge in a unified yet modular framework in which different body plans activate and deactivate different combinations of learned sensorimotor circuits for goal-directed behavior. This allows one part of the controller to be overhauled to better suit new species of designs as they emerge without disrupting the hard-earned knowledge contained within other expert modules. It also allows pretrained expert policies to be directly plugged into the mixture, which can steer evolution into otherwise unexplored areas of latent space containing desired morphological traits. We refer to this process as "evo by demo" and explore how it may be used to guide freeform evolution toward canonical structures defined by the pretrained model. Videos and code can be found at: https://eco-moe.github.io. 

Cooperative perception (CP) enhances situational awareness of connected and autonomous vehicles by exchanging and combining messages from multiple agents. While prior work has explored adversarial integrity attacks that degrade detection accuracy, little is known about CP's robustness against attacks on timeliness (or availability), a safety-critical requirement for autonomous driving. In this paper, we present CP-FREEZER, the first latency attack that maximizes the computation delay of CP algorithms by injecting adversarial perturbation via V2V messages. Our attack resolves several unique challenges, including the non-differentiability of point cloud preprocessing, asynchronous knowledge of the victim’s input due to transmission delays, and uses a novel loss function that effectively maximizes the execution time of the CP pipeline. Extensive experiments show that CP-FREEZER increases end-to-end CP latency by over , pushing per-frame processing time beyond 3 seconds with a 100\% success rate on our real-world vehicle testbed. Our findings reveal a critical threat to the availability of CP systems, highlighting the urgent need for robust defenses. 

Abstract Eutrophication and global warming are key drivers of oxygen loss, also termed deoxygenation, in coastal ecosystems worldwide. Artificial reoxygenation has been suggested as a local or regional solution to increase oxygen concentrations and improve water quality by various parties, including water managers and industry. Three main approaches have been proposed: (1) bubbling with air with the aim to destratify and mix the water column (2) injection of pure oxygen and (3) pumping of oxygenated water to greater water depths (downwelling). In this review, we summarize the results of recent field trials and other implementations of artificial reoxygenation in coastal systems, which, to date, only involve small bays and estuaries. We also discuss potential benefits and risks. While the recent trials indicate that reoxygenation of the water column can be achieved, low oxygen conditions returned rapidly within days to months of discontinuing operations. This illustrates that artificial reoxygenation typically only provides a temporary solution to deoxygenation. Potential side effects of artificial reoxygenation could include enhanced emissions of the greenhouse gas carbon dioxide and, upon bubbling and destratification in shallow waters, also of methane. Additionally, downwelling could lead to warming and an associated increased oxygen demand near the seafloor. Reoxygenation will not necessarily reduce the nutrient availability for phytoplankton, implying that water quality may remain poor. We recommend a careful, case-by-case assessment of the suitability of artificial reoxygenation in coastal systems prior to implementation and monitoring before, during and after each intervention. Any field trials should involve all relevant parties, including scientists and local communities, and results should be reported with full transparency. While in the short-term, artificial reoxygenation may be useful to alleviate oxygen loss in some coastal systems, long-term improvements in the oxygen levels and quality of coastal waters require reductions in nutrient inputs and greenhouse gas emissions.