Multiple Manipulators Research Articles

Frequency-selective acoustic micromanipulation platform

Acoustic micromanipulation methods offer versatile tools ranging from lab-on-chip diagnostic platforms to mobile microrobots. Acoustically oscillating bubble-powered microsystems are gaining increased attention owing to their low cost and convenient operation. To date, no studies have reported on the controllability of the driving frequency on acoustic micromanipulation platforms using Helmholtz resonators. Here, we introduce a Helmholtz resonant cavity into a bubble-powered microsystem to reveal the function of the oscillation modes and frequency selectivity. The Su–Schrieffer–Heeger (SSH) model was utilized to construct a topological interfacial state and bubble-powered working location. We experimentally observed that the microparticles were captured by the oscillation bubble and underwent orbital rotation at the topological frequency. However, the microparticles were tightly bound close to the oscillation bubble area when the driving frequency was outside the topological frequency range. Our acoustic micromanipulation platform implements multiple microparticle manipulation modes based on frequency selectivity that are independent of optical or magnetic properties, which can enrich the micromanipulation toolbox and exhibit enormous application potential in the biomedical field.

Durable semi-interpenetrating polymer network containing dynamic boroxine bonds for multi-shape manipulated deformation

Shape-shifting materials with multiple shape manipulation, integrated with durability, shape reconfiguration, shape memory, and reversible deformation, are in high demand for applications in soft robotics, electrical devices, and other fields, but remain a significant technical challenge. In this study, we present a straightforward approach to develop a semi-interpenetrating polymer network (named PVP-VPBA-PEGDMA) crosslinked with dynamic boroxine bonds to address these challenges. The typical PVP-VPBA2-PEGDMA0.2 exhibits high mechanical strength of 38.5 MPa, exceptional self-healing capability (with up to 89.4 % efficiency), and recyclability, ensuring remarkable durability during shape-changing process. Notably, the PVP-VPBA-PEGDMA can be programmatically transformed from 2D sheets into a variety of 3D shapes using humidity-induced solid-state plasticity, which is facilitated by humidity-sensitive dynamic boroxine bonds. The PVP-VPBA-PEGDMA can be folded like paper through origami techniques, and maintain permanent shape configurations, highly suitable for complex shape morphing applications. Moreover, the PVP-VPBA-PEGDMA, in their temporary geometries, can power a model car using shape memory, demonstrating substantial actuation ability and mechanical performance. Leveraging the asymmetric humidity-triggered behavior of the PVP-VPBA-PEGDMA, various functional humidity-responsive devices have been developed, including a crawling robot, a hat, and a conveyor belt with wet sorting capability. Given the programmable shape reconfiguration, shape memory and reversible shape-changing ability of PVP-VPBA-PEGDMA, it opens up numerous potential applications in soft robotics, flexible electronic devices, and intelligent sensors.

Bi-DexHands: Towards Human-Level Bimanual Dexterous Manipulation.

Achieving human-level dexterity in robotics remains a critical open problem. Even simple dexterous manipulation tasks pose significant difficulties due to the high number of degrees of freedom and the need for cooperation among heterogeneous agents (e.g., finger joints). While some researchers have utilized reinforcement learning (RL) to control a single hand in manipulating objects, tasks that require coordinated bimanual cooperation are still under-explored due to the fewer suitable environments, which can result in difficulties and sub-optimal performance. To address these challenges, we introduce Bi-DexHands, a simulator with two dexterous hands featuring 20 bimanual manipulation tasks and thousands of target objects, designed to match various levels of human motor skills based on cognitive science research. We developed Bi-DexHands in Issac Gym, enabling highly efficient RL training at over 30,000 frames per second using a single NVIDIA RTX 3090. Based on Bi-DexHands, we present a comprehensive evaluation of popular RL algorithms in different settings, including single-agent/multi-agent RL, offline RL, multi-task RL, and meta RL. Our findings show that on-policy algorithms, such as PPO, can master simple manipulation tasks that correspond to those of 48-month-old babies, such as catching a flying object or opening a bottle. Furthermore, multi-agent RL can improve the ability to perform manipulations that require skilled bimanual cooperation, such as lifting a pot or stacking blocks. Despite achieving success in individual tasks, current RL algorithms struggle to learn multiple manipulation skills in most multi-task and few-shot learning scenarios. This highlights the need for further research and development within the RL community.

Generation of High‐Purity Full Poincaré Beam Arrays via Metasurface Integrated Degenerate Cavity

AbstractRecently, full Poincaré beams (FPBs) with spatially variant polarization states spanning the entire surface of the Poincaré sphere have attracted considerable interests. Expanding the FPBs into an array way has the promising applications from multiparticle manipulation to optical communication. However, how to efficiently generate a high‐purity FPB array with intracavity method remains elusive. Here, a metasurface‐integrated degenerate cavity laser is proposed for efficient generation of 10 × 10 FPB array, where a pair of metasurfaces are adopted to independently modulate the circular polarization state while compensating the mode discrepancies within the cavity. Such a design strategy enables a self‐reproducing intracavity optical field, concurrently ensuring the stabilization of intracavity oscillation. It is numerically demonstrated that the purity of orbital angular momentum (OAM) in the first‐order FPB array can approach to 92.89%, with a root‐mean‐square error (RMSE) of polarization ellipse parameters below 8.17%. This work provides a compact and efficient design platform for the generation of a variety of vector beam arrays and may find promising applications in free space communication and multiple particle manipulation.

Distributed bounded finite‐time cooperative control algorithm for multiple nonlinear manipulators

AbstractFor the position synchronization control problem of multiple nonlinear manipulator systems, a bounded distributed cooperative controller is designed in this paper. Based on a fixed‐time position observer, a distributed finite‐time synchronization controller is designed by using the saturated control technique and homogenous system theory. Compared with the existing results, the proposed distributed cooperative control algorithm not only guarantees finite‐time convergence but also satisfies the input saturation. Finally, some numerical simulation results are presented to verify the feasibility of the theory results with six‐degree‐of‐freedom robots.

Bimanual Telemanipulation Framework Utilising Multiple Optically Localised Cooperative Mobile Manipulators

Bimanual manipulation is valuable for its potential to provide robots in the field with increased capabilities when interacting with environments, as well as broadening the number of possible manipulation actions available. However, for a robot to perform bimanual manipulation, the system must have a capable control framework to localise and generate trajectories and commands for each sub-system to allow for successful cooperative manipulation as well as sufficient control over each individual sub-system. The proposed method suggests using multiple mobile manipulator platforms coupled through the use of an optical tracking localisation method to act as a single bimanual manipulation system. The framework’s performance relies on the accuracy of the localisation. As commands are primarily high-level, it is possible to use any number and combination of mobile manipulators and fixed manipulators within this framework. We demonstrate the functionality of this system through tests in a Pybullet simulation environment using two different omnidirectional mobile manipulators, as well a real-life experiment using two quadrupedal manipulators.

Feature Extraction Matters More: An Effective and Efficient Universal Deepfake Disruptor

Face manipulation can modify a victim’s facial attributes, e.g., age or hair color, in an image, which is an important component of DeepFakes. Adversarial examples are an emerging approach to combat the threat of visual misinformation to society. To efficiently protect facial images from being forged, designing a universal face anti-manipulation disruptor is essential. However, existing works treat deepfake disruption as an end-to-end process, ignoring the functional difference between feature extraction and image reconstruction. In this work, we propose a novel F eature- O utput ensemble UN iversal D isruptor (FOUND) against face manipulation networks, which explores a new opinion considering attacking feature-extraction (encoding) modules as the critical task in deepfake disruption. We conduct an effective two-stage disruption process. We first perform ensemble disruption on multi-model encoders, maximizing the Wasserstein distance between features before and after the adversarial attack. Then develop a gradient-ensemble strategy to enhance the disruption effect by simplifying the complex optimization problem of disrupting ensemble end-to-end models. Extensive experiments indicate that one FOUND generated with a few facial images can successfully disrupt multiple face manipulation models on cross-attribute and cross-face images, surpassing state-of-the-art universal disruptors in both success rate and efficiency.

Energy shaping-based consensus control in networked underactuated Euler–Lagrange systems with communication and input delays

This paper aims to solve the problems of energy shaping-based consensus control in networked underactuated Euler–Lagrange systems (NUELSs) in the presence of communication and input delays. By appropriately introducing the auxiliary sliding mode variable describing local communication interaction among agents and the corresponding adaptive velocity observer in energy-shaping framework, two decentralized time-varying consensus tracking schemes are proposed for NUELSs. The developed schemes systematically integrate the energy of three parts: the system model energy of actuated and underactuated components, the energy of the controller dynamics and the energy of the adaptive update law, as the total energy. Then this total energy formulates the resultant Lyapunov function that can fully guarantee the distributed time-varying consensus tracking of NUELSs having time-varying communication and input delays. Furthermore, simulation examples of multiple flexible-joint manipulator systems characterized by NUELSs demonstrate the potential advantages of the proposed consensus schemes in stability, adaptability, and robustness. The effects of communication and input delays on cooperative tracking performance are also numerically investigated for NUELSs.

Optimized leader‐follower consensus control using combination of reinforcement learning and sliding mode mechanism for multiple robot manipulator system

AbstractThis article is to develop an optimized leader‐follower consensus control for multiple robot manipulator system by combining sliding mode control (SMC) and reinforcement learning (RL). The SMC mechanism aims to steer both position and velocity states of multiple manipulator reaching the predefined trajectory. And the RL is designed and performed under identifier‐critic‐actor architecture for the achievement of optimized control performance. Compared to traditional optimal control, this proposed method has two main advantages: (i) the RL updating laws for training both the actor and critic networks are simpler; (ii) the optimized control can not require the complete dynamic knowledge because the adaptive identifier is designed into the RL learning. Consequently, this optimized control method can smoothly steer the multiple robot manipulator system to achieve the leader‐follower consensus. Finally, feasibility of this control method is verified through both theory and simulation.

A biomaterial platform for T cell-specific gene delivery

Efficient T cell engineering is central to the success of CAR T cell therapy but involves multiple time-consuming manipulations, including T cell isolation, activation, and transduction. These steps add complexity and delay CAR T cell manufacturing, which takes a mean time of 4 weeks. To streamline T cell engineering, we strategically combine two critical engineering solutions - T cell-specific lentiviral vectors and macroporous scaffolds - that enable T cell activation and transduction in a simple, single step. The T cell-specific lentiviral vectors (referred to as STAT virus) target T cells through the display of an anti-CD3 antibody and the CD80 extracellular domain on their surface and provide robust T cell activation. Biocompatible macroporous scaffolds (referred to as Drydux) mediate robust transduction by providing effective interaction between naïve T cells and viral vectors. We show that when unstimulated peripheral blood mononuclear cells (PBMCs) are seeded together with STAT lentivirus on Drydux scaffolds, T cells are activated, selectively transduced, and reprogrammed in a single step. Further, we show that the Drydux platform seeded with PBMCs and STAT lentivirus generates tumor-specific functional CAR T cells. This potent combination of engineered lentivirus and biomaterial scaffold holds promise for an effective, simple, and safe avenue for in vitro and in vivo T cell engineering. Statement of significanceManufacturing T cell therapies involves lengthy and labor-intensive steps, including T cell selection, activation, and transduction. These steps add complexity to current CAR T cell manufacturing protocols and limit widespread patient access to this revolutionary therapy. In this work, we demonstrate the combination of engineered virus and biomaterial platform that, together, enables selective T cell activation and transduction in a single step, eliminating multistep T cell engineering protocols and significantly simplifying the manufacturing process.

Soft modularized robotic arm for safe human–robot interaction based on visual and proprioceptive feedback

This study proposes a modularized soft robotic arm with integrated sensing of human touches for physical human–robot interactions. The proposed robotic arm is constructed by connecting multiple soft manipulator modules, each of which consists of three bellow-type soft actuators, pneumatic valves, and an on-board sensing and control circuit. By employing stereolithography three-dimensional (3D) printing technique, the bellow actuator is capable of incorporating embedded organogel channels in the thin wall of its body that are used for detecting human touches. The organogel thus serves as a soft interface for recognizing the intentions of the human operators, enabling the robot to interact with them while generating desired motions of the manipulator. In addition to the touch sensors, each manipulator module has compact, soft string sensors for detecting the displacements of the bellow actuators. When combined with an inertial measurement unit (IMU), the manipulator module has a capability of estimating its own pose or orientation internally. We also propose a localization method that allows us to estimate the location of the manipulator module and to acquire the 3D information of the target point in an uncontrolled environment. The proposed method uses only a single depth camera combined with a deep learning model and is thus much simpler than those of conventional motion capture systems that usually require multiple cameras in a controlled environment. Using the feedback information from the internal sensors and camera, we implemented closed-loop control algorithms to carry out tasks of reaching and grasping objects. The manipulator module shows structural robustness and the performance reliability over 5,000 cycles of repeated actuation. It shows a steady-state error and a standard deviation of 0.8 mm and 0.3 mm, respectively, using the proposed localization method and the string sensor data. We demonstrate an application example of human–robot interaction that uses human touches as triggers to pick up and manipulate target objects. The proposed soft robotic arm can be easily installed in a variety of human workspaces, since it has the ability to interact safely with humans, eliminating the need for strict control of the environments for visual perception. We believe that the proposed system has the potential to integrate soft robots into our daily lives.

Multiple mobile robots motion control based on screw theory for aircraft panel assembly

As a crucial component of large aircraft, the assembly efficiency of aircraft skins directly impacts production efficiency. To achieve efficient manufacturing of aircraft skins, this paper proposes a multiple mobile robot control algorithm based on screw theory. The robot arm is integrated into a rail or AGV to increase its motion space, creating a mobile robot assembly system. To address the redundant degrees of freedom problem caused by the mobile manipulator, this paper adopts the screw theory to describe the motion of multiple robots. Furthermore, to ensure the constraint of the motion between multiple robots, this paper proposes a multi-robot control method based on screw constraint. Rigid body constraints are assigned to the end of each mobile manipulator, and the motion is decomposed to the mobile platform and the robot arm. Finally, the cooperative motion control of multiple mobile manipulators is realized. The proposed algorithm is applied in the multi-mobile manipulator cooperative aircraft panel assembly task, achieving efficient assembly of aircraft panel and long truss.

Dual-band terahertz reflective-mode metasurface for the wavefront manipulation of independent linear and circular polarization waves

Metasurfaces (MSs) are being extensively researched owing to their ability to modulate the polarization and wavefront of electromagnetic (EM) waves in a flexible manner, which usually offer significant advantages including ultra-thinness, low losses, and easy fabrication. However, conventional MSs typically operate well only with a single polarization. Here, we propose a novel design strategy for a terahertz (THz) reflective-mode MS that relies on a single unit-cell arrangement combining propagation phase and geometric phase. Our designed MS can achieve multiple wavefront manipulations in reflection mode, not limited to circular polarization (CP) transformation, but also enabling linear polarization (LP) conversion. The MS we propose consists of a periodic array of bilayered metal patterned resonator structures sandwiched by a dielectric substrate. The metallic resonator is made of the outer single-split-ring (SSR) and C-shaped slot (CSS), inner double-split-ring (DSR), and its complementary structure. With this design, the MS is capable of converting a LP wave to its orthogonal counterpart at lower frequency (f1=0.7THz) after reflection. Additionally, at higher frequency (f2=1.4THz), the proposed MS can also convert the right-handed CP (RCP) to left-handed CP (LCP) upon reflection or vice versa. The 2π phase full coverage of the orthogonal LP and CP waves can be achieved independently and simultaneously by adjusting the opening and orientation angles of the SSR based on propagation phase, and orientation angle of the DSR based on geometric phase. We numerically demonstrate beam deflection, planar focusing, and the vortex beam for both reflected orthogonal LP and CP waves with three representative MSs to provide proof of concept. These findings reveal the great potential for multifunctional devices for dual-polarization in imaging and communication systems.

Online and Scalable Motion Coordination for Multiple Robot Manipulators in Shared Workspaces

Multi-robot motion coordination is essential to make robots work safely and efficiently in a shared workspace, ensuring task completion while avoiding interference between each other’s motions. Achieving online replanning and scaling to large numbers of robots are particularly challenging. In this paper, we present a motion coordination method for multiple robot manipulators with given targets. The approach separates the problem into a global coordination stage and a local trajectory replanning stage. Online reactivity and scalability are achieved by means of coordination in a reduced space and efficient trajectory replanning. We first introduce the method for systems with two robots, then extend it to systems with an arbitrary number of robots. The method determines whether kinematically-feasible and non-interfering trajectories leading all the robots to their targets exist. The approach generates time-efficient trajectories if solutions exist, or provides information for switching targets if solutions cannot be found. We show formally and empirically that the method has low computational overhead and scales quadratically with the number of robots. Experiments are conducted with up to three real 7-DOF robots and up to ten simulated robots. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —In underground mining, a key process is that of tunneling, i.e., drilling and blasting rock to excavate tunnels that lead to sources of ore. Drilling is carried out by rigs equipped with multiple robotic arms. A key factor affecting the efficiency of tunneling is the time to completion of drilling operations. In current industrial practice, these operations are carried out manually by an operator steering the arms on the drill rig. Time to completion can be drastically reduced if the arms could operate concurrently, intelligently optimizing and coordinating their motions. However, existing methods for multi-arm motion coordination are inadequate, as they fail to cater to one or more of the following real-world constraints: several robot arms work in close proximity and their workspaces are overlapping; task completion times are uncertain due to rock density and drill bit breakage; and contingencies such as unexpected delays in motion or stops sometimes happen. These constraints exist also in other industrial applications, like manufacturing and assembly. This paper proposes a framework which enables multiple high-DOF robot manipulators to safely and efficiently work in a shared workspace. The method allows to adjust robot trajectories while robots move, and is shown to scale well with the number of robots. Motions are generated and adjusted by time-optimal trajectory planning, whose computational overhead is small enough for online operation and benefits task completion efficiency, safety, and productivity. Experiments using three 7-DOF robots suggest that this approach is practically feasible, and simulations with up to 10 robots attest to its scalability.

Observer-Based Adaptive Prescribed- Time H ∞ Coordinated Control for Multiple Robot Manipulators With Prescribed Performance and Input Quantization

Observer-Based Adaptive Prescribed- Time <i>H</i> _∞ Coordinated Control for Multiple Robot Manipulators With Prescribed Performance and Input Quantization

Adaptive Cooperative Output Regulation for Multiple Flexible Manipulators

Adaptive Cooperative Output Regulation for Multiple Flexible Manipulators

Adaptive backstepping asymptotic consensus tracking control of multiple uncertain manipulators with disturbances

AbstractThis article studies the asymptotic consensus of multiple robots with uncertain parameters and external disturbances. A novel asymptotic tracking command filtered adaptive consensus control approach is proposed. The command filters avoid the differentials of virtual control functions in backstepping, meanwhile, the compensation mechanism is used to compensate filtering errors. Compared with current command filtered control algorithms for the consensus control of multiple robots, the consensus tracking errors can asymptotically converge to origin under the given approach, moreover, the design process is simplified by designing only one adaptive parameter for each manipulator. A simulation example verifies the given approach.

Evaluating Machine Learning Methods of Analyzing Multiclass Metabolomics.

Multiclass metabolomic studies have become popular for revealing the differences in multiple stages of complex diseases, various lifestyles, or the effects of specific treatments. In multiclass metabolomics, there are multiple data manipulation steps for analyzing raw data, which consist of data filtering, the imputation of missing values, data normalization, marker identification, sample separation, classification, and so on. In each step, several to dozens of machine learning methods can be chosen for the given data set, with potentially hundreds or thousands of method combinations in the whole data processing chain. Therefore, a clear understanding of these machine learning methods is helpful for selecting an appropriate method combination for obtaining stable and reliable analytical results of specific data. However, there has rarely been an overall introduction or evaluation of these methods based on multiclass metabolomic data. Herein, detailed descriptions of these machine learning methods in multiple data manipulation steps are reviewed. Moreover, an assessment of these methods was performed using a benchmark data set for multiclass metabolomics. First, 12 imputation methods for imputing missing values were evaluated based on the PSS (Procrustes statistical shape analysis) and NRMSE (normalized root-mean-square error) values. Second, 17 normalization methods for processing multiclass metabolomic data were evaluated by applying the PMAD (pooled median absolute deviation) value. Third, different methods of identifying markers of multiclass metabolomics were evaluated based on the CWrel (relative weighted consistency) value. Fourth, nine classification methods for constructing multiclass models were assessed using the AUC (area under the curve) value. Performance evaluations of machine learning methods are highly recommended to select the most appropriate method combination before performing the final analysis of the given data. Overall, detailed descriptions and evaluation of various machine learning methods are expected to improve analyses of multiclass metabolomic data.

Generation and screening of mouse models of ADRD with multiple genetic manipulations to model dementia with mixed brain pathologies

AbstractBackgroundAlzheimer’s disease‐related dementias (ADRD) encompass a wide range of neurodegenerative diseases including Alzheimer’s disease, vascular dementia, frontotemporal dementia (FTD), Lewy body dementia (LBD), and mixed dementia [1‐4]. ADRDs are distinguished by the presence of pathologies including amyloid, tau, TDP‐43 inclusions, alpha‐synuclein aggregates, cerebral amyloid angiopathy, and/or hippocampal sclerosis [4‐9]. Consequently, the heterogeneity of ADRDs presents challenges for animal modeling of disease mechanisms. Our laboratory has demonstrated that genetic diversity in disease modeling is critical to recapitulate the molecular and cognitive features of AD [1,9, 10]. Here, we introduce multiple genetic manipulations to develop novel translational models of ADRD.MethodPrecision models of ADRD on multiple genetic backgrounds were generated via combinatorial CRISPR editing to knock‐in human disease‐causing mutations onto the C57BL/6J (B6) strain and crossed to either B6, DBA/2J (D2), FVB, or WSB to generate a set of 20 unique ADRD models (Fig.1). The effect of homozygous mutations were evaluated on the B6 background. Each model was assessed using a high‐capacity multi‐domain go/no‐go screen to prioritize candidate strains for future ADRD modeling of drug discovery, preclinical testing, and disease pathology.ResultAssays included frailty, open field, y‐maze, rotarod, and contextual fear conditioning. Models with one or more late‐onset behavioral deficits (12 mo) were a “go”; whereas, early behavioral deficits (6 and 9 mo), aggression, high attrition, and low fecundity and additional factors were considered unsuitable (“no‐go”). We identified several promising models of ADRD that exhibited late‐onset frailty and motor deficits as function of mutation burden and genetic background (e.g., App/Psen1/Tardbp/Mapt on the D2 background). The FVB background strain exhibited early frailty deficits (6 mo), and thus considered unsuitable (“no‐go”). The WSB background strain exhibited reserve across multiple frailty and behavioral assays up to 12 mo.ConclusionOur results demonstrate that combining ADRD mutations in genetically distinct backgrounds is a promising strategy to generate new mouse models that better model the human mid and late‐life onset of deficits associated with ADRD. The resulting model(s) will be translationally relevant and provide important insight into disease mechanisms for the future of ADRD research and drug discoveries.

Lipid Nanoparticle-Mediated Gene Editing of Human Primary T Cells and Off-Target Analysis of the CRISPR-Cas9 Indels

Background and Aims: The CRISPR/Cas9 system has emerged as a powerful tool for gene editing of primary cells. In our previous work, we demonstrated a novel lipid nanoparticle (LNP) reagent for the multi-step engineering of gene-edited CAR T cells. We showed high cell viabilities and potent CAR-T mediated killing, even after multiple genetic manipulations. Here, we extend this work by assessing potential off-target editing effects in both LNP-treated T cells and T-cells where the CRISPR reagents were delivered by electroporation. Further, we evaluated multiple Cas9 variants and guide RNA targets. Methods: LNPs encapsulating wild type or high fidelity S.p. Cas9 mRNA and various TRAC and CD52 targeted guide RNAs (sgRNAs) were produced using our scalable NanoAssemblrTM microfluidics platform. Concurrently, electroporation was performed to deliver equivalent cargoes. Purified primary T cells were cultured, activated, and expanded in serum-free media in plates, flasks, or small bioreactors. The LNPs were added to cells by direct addition for gene editing. Gene expression and cell viability were measured using flow cytometry or colorimetric assays. Multi-target performance of CRISPR-Cas9 editing was evaluated through rhAmpSeq-based targeted next-generation sequencing (NGS) and indels analyzed for with CRISPRAltRations. Results: TCR or CD52 targeted Cas9 mRNA-LNP addition or electroporation yielded high single and double knockout efficiencies. High-throughput NGS analysis showed strong agreeance to flow cytometry for on-target analysis. We tested a range of sgRNA targets, wild-type and high-fidelity Cas9 mRNAs, and determined off-target editing for all targets and variants investigated. Similar results were obtained when comparing different LNP batch sizes (micro to milligram RNA) and cell culture vessels (0.1 to 45 million cells), demonstrating scalability of both LNP production and cell treatment. Conclusions: The results from this study further support the utility RNA-LNPs for the genetic engineering of primary T cells. The simple and gentle nature of LNP cell treatment allows for multiple genetic engineering steps for simultaneous expression and deletion of proteins for the next cell therapies. These LNPs can be easily manufactured from small-scale screening of RNA libraries to rapid scale-up for clinical translation.