An autonomous, solar-powered robot designed to travel along suspended wires for long-term, non-invasive environmental monitoring in hard-to-reach natural areas.

Organoid/brain slice immobilization for microelectrode arrays (MEAs) and organoid-on-chip platforms have traditionally depended on hydrogels, harp-style grids, or microfluidic confinement, each with its own set of pros and cons with respect to stability, standardization, and impact on electrophysiology. Hydrogels (e.g., Polyethylene glycol or PEG, extracellular matrix like Matrigel) are widely used to immobilize 3D neural tissues on MEAs. These are known to swell, drift, and alter mechanical microenvironments, which in turn modulate network firing, synchrony, and bursting behavior. Mechanical retention via harp slice grids or similar harp devices is a long-standing practice in acute brain slice and organoid electrophysiology. These devices are typically standardized, fragile, and poorly matched to diverse well and tissue geometries. ​Microfluidic organoid chips and specialized 3D MEAs (e.g., e-Flower, organoid-on-chip platforms) have recently emerged to enable hydrogel-free trapping/encapsulation of organoids for imaging and recordings, but they often require bespoke chip designs and overly complex flow control setups. There is a lack of geometry-agnostic devices for mechanically immobilizing diverse organoids on commercial MEAs that feature consistent stability, uniform and/or tailored contact, and with minimal perturbation of electrophysiological readouts.

Researchers at the University of California, Davis have developed a system of reusable, sterilizable 3D-printed surgical tools that enables safe, precise intraoperative deployment of Neuropixels probes within standard neurosurgical workflows.

The deployment of robotic systems in real-world environments is often limited by the "sim-to-real gap," where policies trained in digital simulations fail to account for the complex, multisensory feedback of physical reality. Researchers at UC Berkeley have developed a novel method for training multimodal sim-to-real robot policies by integrating generative audio models with traditional physics-based simulators. This framework uses a generative model to synthesize realistic audio data that corresponds to simulated physical interactions, creating a rich, multimodal dataset for policy learning. By training on both simulated physics and generated sensory data, the system enables robots to develop more robust and adaptive behaviors that translate seamlessly from virtual training environments to complex real-world tasks.

Brief description not available

Traditional environmental policymaking often struggles to efficiently target interventions to achieve multiple, complex air quality goals simultaneously across a geographic area. This innovation, developed by UC Berkeley researchers, addresses this challenge by providing a sophisticated, multi-objective optimization method for targeted reduction of air pollution. The method generates a comprehensive mitigation pathway by integrating several modules: a forward module to model pollutant concentrations, a target concentration surface that defines the policy goals, a prioritization module to assess uncertainty and importance via a prioritization covariance matrix, and a Bayesian inversion module to estimate optimum emissions required to meet the target. This systematic, data-driven approach culminates in a mitigation pathway that guides the performance of specific pollution control measures, offering a significant advantage over conventional, less targeted policy-making by ensuring resources are directed where they will have the maximum environmental impact.

UC Berkeley researchers have developed a novel dispersion system for agricultural and environmental payloads, including seeds, soil amendments, miniature soil sensors, and so forth. Dispersive packages are biodegradable and biomimetically designed with similarities to natural seeds. Aerodynamic properties control large-area dispersions, while importantly, tunable gyroscopic properties are programmed for penetration parameters, such as depth, upon impact. Payload distribution can be fine-tuned accounting for local soil moisture and grain-size.

Researchers at the University of California, Davis have developed tactile feedback systems that enhance spatial and sensory resolution in sensor arrays through unique signal modulation techniques.

Researchers at the University of California, Davis have developed an approach to enhance autonomous vehicle path prediction through efficient information sharing and distributed learning among AI agents.

A user-guided iterative pipeline that significantly improves the reliability and quality of code generated by Large Language Models (LLMs) for industrial control systems (ICS).

Current sustainable building materials often lack the high structural strength needed for demanding applications, limiting their use in load-bearing construction. Addressing this opportunity, UC Berkeley researchers have developed SEA-BOARD, a novel structural panel fabricated from marine-derived polysaccharides. This innovation utilizes a proprietary, stepwise process involving polysaccharide extraction, nanofiber alignment, and thermal densification to configure the macroalgal biomass into a high-strength, hot-pressed panel. This engineered material is structurally superior and potentially more environmentally sustainable than many traditional wood-based or synthetic alternatives.

An innovative, AI-driven approach for non-intrusive analysis and defect detection in multi-tiered chiplets, enhancing microelectronics packaging.

Researchers at the University of California, Davis have developed a technology that introduces a highly adaptable, lightweight robotic end effector designed for complex manipulation tasks in automation.

A novel motorized tool designed to precisely deliver retinal tissue during transplantation, enhancing outcomes for patients with retinal degeneration.

Brief description not available

Brief description not available

ROMANUS is a cutting-edge methodology designed to enhance the performance and robustness of latency-critical, real-time intelligent systems through dynamic neural architectures.

This technology enables precise, selective manipulation of magnetically barcoded materials, distinguishing them from background magnetic materials

Advances in biological research have been greatly influenced by the development of organoids, a specialized form of 3D cell culture. Created from pluripotent stem cells, organoids are effective in vitro models in replicating the structure and progression of organ development, providing an exceptional tool for studying the complexities of biology. Among these, cerebral cortex organoids (hereafter "organoid") have become particularly instrumental in providing valuable insights into brain formation, function, and pathology. Despite their potential, organoid experiments present several challenges. Organoids require a rigorous, months-long developmental process, demanding substantial resources and meticulous care to yield valuable data on aspects of biology such as neural unit electrophysiology, cytoarchitecture, and transcriptional regulation. Traditionally the data has been difficult to collect on a more frequent and consistent basis, which limits the breadth and depth of modern organoid biology. Generating and measuring organoids depend on media manipulations, imaging, and electrophysiological measurements. Historically are labor- and skill-intensive processes which can increase risks associated with experimental validity, reliability, efficiency, and scalability.

Advances in biological research have been greatly influenced by the development of organoids, a specialized form of 3D cell culture. Created from pluripotent stem cells, organoids are effective in vitro models in replicating the structure and progression of brain development, providing an exceptional tool for studying the complexities of biology. Among these, cortical organoids, comprising in part of neurons, have been instrumental in providing early insights into brain formation, function, and pathology. Functional characteristics of cortical organoids, such as cellular morphology and electrophysiology, provide physiological insight into cellular states and are crucial for understanding the roles of cell types within their specific niches. And while progress has been made studying engineered neuronal systems, decoding the functional properties of neuronal networks and their role in producing behaviors depends in part on recognizing neuronal cell types, their general locations within the brain, and how they connect.

Brief description not available

Advances in biological research have been greatly influenced by the development of organoids, a specialized form of 3D cell culture. Created from pluripotent stem cells, organoids are effective in vitro models in replicating the structure and progression of organ development, providing an exceptional tool for studying the complexities of biology. Among these, cerebral cortex organoids (hereafter “organoid”) have become particularly instrumental in providing valuable insights into brain formation, function, and pathology. Despite their potential, organoid experiments present several challenges. Organoids require a rigorous, months-long developmental process, demanding substantial resources and meticulous care to yield valuable data on aspects of biology such as neural unit electrophysiology, cytoarchitecture, and transcriptional regulation. Traditionally the data has been difficult to collect on a more frequent and consistent basis, which limits the breadth and depth of modern organoid biology. Generating and measuring organoids depend on media manipulations, imaging, and electrophysiological measurements. Historically these are labor- and skill-intensive processes which can increase risks associated with known human error and contamination.

Bridging the gap between linguistic reasoning and physical execution, UC Berkeley researchers have developed a method to enable robotic devices to predict complex actions using in-context learning (ICL). By leveraging the inherent reasoning capabilities of Large Language Models (LLMs), this approach allows a robot to translate natural language instructions into sequential motor actions without the need for task-specific fine-tuning or intensive retraining. The system allows the robot to generalize to new, unseen tasks on the fly. This breakthrough shifts robot programming away from rigid coding toward a more flexible, intuitive interaction where the machine "understands" the intended goal by drawing parallels from the provided examples.

Bridging the gap between a language model’s next-word prediction and physical robot control, researchers at UC Berkeley have developed LLARVA (Large Language model for Robotic Vision and Action). This model utilizes a novel vision-action instruction tuning method that allows a robotic device to handle various tasks and environments without task-specific fine-tuning.

Advancing the field of robotic agility, this technology treats the complex challenge of bipedal balance and movement as a generative sequence problem. By framing physical movement similarly to language modeling, UC Berkeley researchers have developed a system where a humanoid robot predicts its next motor action as a "next token" based on a vast history of sensorimotor trajectories. The model is trained on diverse data, including real-world robotic walks and simulated movements, allowing it to anticipate the necessary joint adjustments and equilibrium shifts in real-time. This approach enables the robot to navigate uneven terrain and respond to external perturbations with a level of fluidity and adaptability that traditional, rigidly programmed control laws often struggle to achieve.