Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “REINFORCEMENT”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,657 records · Page 92

Influence of Core and Hexapod Geometry, and Local Reinforcement on the Performance of Ultra-Lightweight ULE Mirror

The Habitable Exoplanet Observatory (HabEx) mission has unique optical performance requirements which drive the mirror design process beyond the traditional criteria. While mass and stiffness are still important, the response to inertia loading (expressed in terms of Zernike coefficients) to omni-directional excitation dominates the effort. While a Zerodur mirror is the current baseline, as mass budgets change, a ULE design is being studied as a potential alternative. This trade study looked at over 264 design variations using the Arnold Mirror Modeler and ANSYS(c) to investigate the influence of various design elements, including: substrate thickness, core cell size, hexapod geometry and local reinforcement. Design 'goodness' was evaluated based on the mirror's inertial deformation response to omni-directional input. This response was calculated via RSSing Zernike polynomial responses to (XYZ) accelerations.

Space Telescope Technology↗

The Effects of Fiber Orientation and Adhesives on Tensile Properties of Carbon Fiber Reinforced Polymer Matrix Composite with Embedded Nickel-Titanium Shape Memory Alloys

Tensile tests of Nickel-titanium (NiTi) shape memory alloys (SMA) embedded within carbon fiber reinforced polymer matrix composite (CFRP/PMC) laminates were evaluated with simultaneous monitoring of modal acoustic emissions (MAE). Three different layup configurations utilizing two different thin film adhesives were applied to bond the materials. Ultimate tensile strengths, strains, and moduli were obtained along with cumulative AE energy of events and specimen failure location. Scanning electron microscopy was used to examine the break areas of the specimens post-test. Microscopy was used to validate failure locations revealed from MAE analysis. A unique finding within this research showed that 90° plies in the outer ply gave the strongest acoustic signals as well as the cleanest fracture of the specimens tested. Overlapping 0° ply layers surrounding the SMA was found to be the best scenario to prevent failure of the specimen itself.

Derek Quade↗

Surface Characterization of Carbon Fiber Reinforced Polymers by Picosecond Laser Induced Breakdown Spectroscopy

Adhesive bonding of composite materials requires reliable monitoring and detection of surface contaminants as part of a vigorous quality control process to assure robust and durable bonded structures. Surface treatment and effective monitoring prior to bonding are essential in order to obtain a surface which is free from contaminants that may lead to inferior bond quality. In this study, the focus is to advance the laser induced breakdown spectroscopy (LIBS) technique by using pulse energies below 100 μJ (μLIBS) for the detection of low levels of silicone contaminants in carbon fiber reinforced polymer (CFRP) composites. Various CFRP surface conditions were investigated by LIBS using ∼10 ps, 355 nm laser pulses with pulse energies below 30 μJ. Time-resolved analysis was conducted to optimize the gate delay and gate width for the detection of the C I emission line at 247.9 nm to monitor the epoxy resin matrix of CFRP composites and the Si I emission line at 288.2 nm for detection of silicone contaminants in CFRP. To study the surface sensitivity to silicone contamination, CFRP surfaces were coated with polydimethylsiloxane (PDMS), the active ingredient in many mold release agents. The presence of PDMS was studied by inspecting the Si I emission lines at 251.6 nm and 288.2 nm. The measured PDMS areal densities ranged from 0.15 to 2 μg/cm(sup 2). LIBS measurements were performed before and after laser surface ablation. The results demonstrate the successful detection of PDMS thin layers on CFRP using picosecond μLIBS.

Rodolfo Ledesma↗

High Strain Rate Response of Adhesively Bonded Fiber-Reinforced Composite Joints – A Computational Study to Guide Experimental Design

Adhesively bonded carbon fiber-reinforced epoxy composite laminates are widely used in aerospace applications. During a high energy impact event, these laminates are often subjected to high strain rate loading. However, the influence of high strain rate loading on the response of these composite joints is not well understood. Computational finite element (FE) modeling and simulations are conducted to guide the design of high strain rate experiments. Two different experimental designs based on split Hopkinson bar were numerically modeled to simulate Mode I and Mode II types loading in the composite. In addition, the computational approach adopted in this study helps in understanding the high strain rate response of adhesively bonded composite joints subjected to nominally Mode I and Mode II loading. The modeling approach consists of a ply-level 3D FE model, a progressive damage constitutive model for the composite material behavior and a cohesive tie-break contact element for interlaminar delamination.

Ravindran, Suraj↗

Autonomous Spacecraft Attitude Control Using Deep Reinforcement Learning

While machine learning and spacecraft autonomy continue to gain research interest, significant work remains to be done in efficiently applying modern machine learning techniques to problems in space ight. This study presents a framework for deriving a discrete neural spacecraft attitude controller using reinforcement learning, a paradigm of machine learning, without the need for high-performance computing. The developed attitude controller is an approximately time-optimal solution to a highly constrained control problem, able to achieve well above industry-standard pointing accuracies. Control examples are also presented of the agent performing large-angle spacecraft slews in the developed simulation environment and future extensions of this work are discussed.

ATAP↗

Exploring Transfers Between Earth-Moon Halo Orbits via Multi-Objective Reinforcement Learning

Multi-Reward Proximal Policy Optimization, a multi-objective deep reinforcement learning algorithm, is used to examine the design space of low-thrust trajectories for a SmallSat transferring between two libration point orbits in the Earth-Moon system. Using Multi-Reward Proximal Policy Optimization, multiple policies are simultaneously and efficiently trained on three distinct trajectory design scenarios. Each policy is trained to create a unique control scheme based on the trajectory design scenario and assigned reward function: a unique combination of weights scaling competing objectives that guide the spacecraft to the target mission orbit, incentivize faster flight times, and penalize propellant mass usage. Then, the policies are evaluated on the same set of perturbed initial conditions in each scenario to generate the propellant mass usages, flight times, and state discontinuities from a reference trajectory for each control scheme. This solution space of low-thrust trajectories for a SmallSat is used to examine the multi-objective trade space for the trajectory design scenario. By autonomously constructing the solution space, insights into the required propellant mass, flight time, and transfer geometry are rapidly achieved.

Christopher J Sullivan↗

Carbon Fiber-Carbon Nanotube Yarn Hybrid Reinforcement

This presentation describes the background and motivation behind the development of the carbon fiber-carbon nanotube yarn hybrid reinforcement. The talk will be presented as part of NASA's monthly Tech Transfer webinars.

nanotubes↗

MULTI-OBJECTIVE REINFORCEMENT LEARNING FOR LOW-THRUST TRANSFER DESIGN BETWEEN LIBRATION POINT ORBITS

Multi-Reward Proximal Policy Optimization (MRPPO) is a multi-objective reinforcement learning algorithm used to construct low-thrust transfers between periodic orbits in multi-body systems. Previous implementations of MRPPO have relied on a predefined reference transfer to successfully train each policy. In this paper, an algorithmic modification labeled the ‘moving reference’, is introduced to autonomously construct these reference trajectories during training. With this modification, MRPPO is used to recover various low-thrust transfers between two periodic orbits in the Earth-Moon circular restricted three-body problem to solve a multi-objective optimization problem. These results are then compared with the solutions recovered via a traditional optimization formulation.

algorithm↗

Artificial Generation of 2-D Fiber Reinforced Composite Microstructures with Statistically Equivalent Features

Fiber reinforced composites are used widely for their high strength and low weight advantages in various aerospace and automotive applications. While their use may be sought after, modeling of these material requires increasing fidelity at the lower scales to capture accurate material behavior under loading. The first steps in creating statistically equivalent models to real life cases is developing a method of rapid evaluation and artificial microstructure generation. The outlined work is capable of tracking microscale fiber positions and determining regions of localized volume fraction extrema (high and low end). Groupings of high and low volume fraction regions are called clusters and their geometry is used to characterize the microstructure. These cluster features can be evaluated for both artificial models and actual scans, allowing correlation to be established which can ultimately be used to regenerate statistically equivalent models. The results of this work show that if one feature is to be correlated, a model can be generated which matches almost exactly. But once more features are equally taken into account, the regeneration loses accuracy.

micromechanics↗

Dynamic Spectrum Allocation in Urban Air Transportation System via Deep Reinforcement Learning

The emerging concepts of Urban Air Mobility (UAM) and Advanced Air Mobility (AAM) open a new paradigm for urban air transportation. A big challenge is that these new aerial vehicles will quickly saturate the already crowded aviation spectrum, which is an essential resource to ensure reliable communications for safe operations. In this paper, we consider an air transportation system where multiple aerial vehicles are operated to transport passengers or cargo from different sources to destinations along their pre-defined paths. During the flight, the minimum communication Quality of Service (QoS) requirement must be achieved to ensure flight safety. Our objective is to minimize the average mission completion time by jointly optimizing the velocity selection and spectrum allocation for all aerial vehicles. We formulate the optimization problem as a multi-stage Markov Decision Process (MDP) where the optimization variables are coupled together. A multi-agent Deep Reinforcement Learning (DRL) based solution is proposed where Value Decomposition Networks (VDN) algorithm is utilized to take discrete actions. Additionally, we propose a heuristic greedy algorithm as a baseline solution. Simulation results show that our learning based solution outperforms the heuristic greedy algorithm and another Orthogonal Multiple Access (OMA) solution in minimizing the mission completion time.

Ruixuan Han↗

MULTI-OBJECTIVE REINFORCEMENT LEARNING FOR LOW-THRUST TRANSFER DESIGN BETWEEN LIBRATION POINT ORBITS

Multi-Reward Proximal Policy Optimization (MRPPO) is a multi-objective reinforcement learning algorithm used to construct low-thrust transfers between periodic orbits in multi-body systems. Previous implementations of MRPPO have relied on a predefined reference transfer to successfully train each policy. In this paper, an algorithmic modification labeled the ‘moving reference’, is introduced to autonomously construct these reference trajectories during training. With this modification, MRPPO is used to recover various low-thrust transfers between two periodic orbits in the Earth-Moon circular restricted three-body problem to solve a multi-objective optimization problem. These results are then compared with the solutions recovered via a gradient descent optimization scheme to validate the performance of MRPPO with the moving reference modification.

Christopher J. Sullivan↗

Dynamic Spectrum Allocation in Urban Air Transportation System via Deep Reinforcement Learning

The emerging concepts of Urban Air Mobility (UAM) and Advanced Air Mobility (AAM) open a new paradigm for urban air transportation. A big challenge is that these new aerial vehicles will quickly saturate the already crowded aviation spectrum, which is an essential resource to ensure reliable communications for safe operations. In this paper, we consider an air transportation system where multiple aerial vehicles are operated to transport passengers or cargo from different sources to destinations along their pre-defined paths. During the flight, the minimum communication Quality of Service (QoS) requirement must be achieved to ensure flight safety. Our objective is to minimize the average mission completion time by jointly optimizing the velocity selection and spectrum allocation for all aerial vehicles. We formulate the optimization problem as a multi-stage Markov Decision Process (MDP) where the optimization variables are coupled together. A multi-agent Deep Reinforcement Learning (DRL) based solution is proposed where Value Decomposition Networks (VDN) algorithm is utilized to take discrete actions. Additionally, we propose a heuristic greedy algorithm as a baseline solution. Simulation results show that our learning based solution outperforms the heuristic greedy algorithm and another Orthogonal Multiple Access (OMA) solution in minimizing the mission completion time.

Ruixuan Han↗

MULTI-OBJECTIVE REINFORCEMENT LEARNING FOR LOW-THRUST TRANSFER DESIGN BETWEEN LIBRATION POINT ORBITS

Multi-Reward Proximal Policy Optimization (MRPPO) is a multi-objective reinforcement learning algorithm used to construct low-thrust transfers between periodic orbits in multi-body systems. Previous implementations of MRPPO have relied on a predefined reference transfer to successfully train each policy. In this paper, an algorithmic modification labeled the ‘moving reference’, is introduced to autonomously construct these reference trajectories during training. With this modification, MRPPO is used to recover various low-thrust transfers between two periodic orbits in the Earth-Moon circular restricted three-body problem to solve a multi-objective optimization problem. These results are then compared with the solutions recovered via a gradient descent optimization scheme to validate the performance of MRPPO with the moving reference modification

Christopher John Sullivan↗

Exploring the Low-Thrust Transfer Design Space in an Ephemeris Model via Multi-Objective Reinforcement Learning

Multi-Reward Proximal Policy Optimization (MRPPO) is a multi-objective reinforcement learning algorithm used to train multiple policies to uncover solutions within a multi-objective solution space. MRPPO is used in this paper to train policies to construct low-thrust transfers for a SmallSat from the vicinity of !2 to an !5 short period orbit in the Sun-Earth-Moon system. First, the policies are trained in this scenario in the circular restricted three-body problem. This information is used to initialize the policies before training in a higher-fidelity ephemeris model; a process known as transfer learning. The recovered segments of the solution space will be compared to fundamental dynamical structures to both examine the results of MRPPO in this complex design scenario and explore the effectiveness of transfer learning.

Christopher J Sullivan↗

Exploring the Low-Thrust Transfer Design Space in an Ephemeris Model via Multi-Objective Reinforcement Learning

Multi-Reward Proximal Policy Optimization (MRPPO) is a multi-objective reinforcement learning algorithm used to train multiple policies to uncover solutions within a multi-objective solution space. MRPPO is used in this paper to train policies to construct low-thrust transfers for a SmallSat from the vicinity of L2 to an L5 short period orbit in the Sun-Earth-Moon system. First, the policies are trained in this scenario in the circular restricted three-body problem. This information is used to initialize the policies before training in a higher-fidelity ephemeris model; a process known as transfer learning. The recovered segments of the solution space will be compared to fundamental dynamical structures to both examine the results of MRPPO in this complex design scenario and explore the effectiveness of transfer learning.

Christopher J. Sullivan↗

An Online Reinforcement Learning Controller Design For Mars Ascent Vehicle

This paper presents a neural network (NN) approximator-based online reinforcement learning (ORL) controller design for Mars Ascent Vehicle (MAV) under parametric variation and significant external disturbances. The ORL controller, which does not require any offline training, involves two NNs where an action NN produces optimal short-term control performance while a critic NN evaluates the performance of the action NN using an approximated cost function. The simulation example with comparisons against baseline Proportional-Integral-Derivative (PID) and gain scheduled pole-placement PID (GS-PP-PID) controllers show the proposed controller’s effectiveness and robustness under parametric variation and high external disturbances.

Han Woong Bae↗

Exploring Transfers Between Earth-Moon Halo Orbits via Multi-Objective Reinforcement Learning

Multi-Reward Proximal Policy Optimization, a multi-objective deep reinforcement learning algorithm, is used to examine the design space of low-thrust trajectories for a SmallSat transferring between two libration point orbits in the Earth- Moon system. Using Multi-Reward Proximal Policy Optimiza- tion, multiple policies are simultaneously and efficiently trained on three distinct trajectory design scenarios. Each policy is trained to create a unique control scheme based on the trajectory design scenario and assigned reward function: a unique combination of weights scaling competing objectives that guide the spacecraft to the target mission orbit, incentivize faster flight times, and penalize propellant mass usage. Then, the policies are evaluated on the same set of perturbed initial conditions in each scenario to generate the propellant mass usages, flight times, and state discontinuities from a reference trajectory for each control scheme. This solution space of low-thrust trajectories for a SmallSat is used to examine the multi-objective trade space for the trajectory design scenario. By autonomously constructing the solution space, insights into the required propellant mass, flight time, and transfer geometry are rapidly achieved.

Mashiku, Alinda K.↗

Scheduling the NASA Deep Space Network with Deep Reinforcement Learning

With three complexes spread evenly across the Earth, NASA’s Deep Space Network (DSN) is the primary means of communications as well as a significant scientific instrument for dozens of active missions around the world. A rapidly rising number of spacecraft and increasingly complex scientific instruments with higher bandwidth requirements have resulted in demand that exceeds the network’s capacity across its 12 antennae. The existing DSN scheduling process operates on a rolling weekly basis and is time-consuming; for a given week, generation of the final baseline schedule of spacecraft tracking passes takes roughly 5 months from the initial requirements submission deadline, with several weeks of peer-to-peer negotiations in between. This paper proposes a deep reinforcement learning (RL) approach to generate candidate DSN schedules from mission requests and spacecraft ephemeris data with demonstrated capability to address real-world operational constraints. A deep RL agent is developed that takes mission requests for a given week as input, and interacts with a DSN scheduling environment to allocate tracks such that its reward signal is maximized. A comparison is made between an agent trained using Proximal Policy Optimization and its random, untrained counterpart. The results represent a proof-of-concept that, given a well-shaped reward signal, a deep RL agent can learn the complex heuristics used by experts to schedule the DSN. A trained agent can potentially be used to generate candidate schedules to bootstrap the scheduling process and thus reduce the turnaround cycle for DSN scheduling.

Wilson, Brian↗