Lockheed Martin Advanced Technology Laboratories

Traditional artificial neural networks take inspiration from biological networks, using layers of neuron-like nodes to pass information for processing. More realistic models include spiking in the neural network, capturing the electrical characteristics more closely. However, a large proportion of brain cells are of the glial cell type, in particular astrocytes which have been suggested to play a role in performing computations. Here, we introduce a modified spiking neural network model incorporating artificial astrocytes and assess their impact on learning. We implement the network as a liquid state machine and task the network with performing a chaotic time-series prediction task. We varied the number and ratio of artificial neurons and astrocytes in the network to examine the latter units' effect on learning. We show that networks combining both neurons and astrocytes together, as opposed to neural- and astrocyte-only networks, are critical for driving learning. Interestingly, we found that the highest learning rate was achieved when the ratio between artificial astrocytes and neurons was roughly 2:1, mirroring some estimates of the ratio of biological astrocytes to neurons. Our results demonstrate that incorporating artificial astrocytes which represent information across longer timescales can alter the learning rates of neural networks, and the proportion of astrocytes to neurons should be tuned appropriately to a given task.

Long-term ship trajectory prediction is a fundamental capability for maritime safety and autonomous navigation. While recent Transformer-based architectures have improved forecasting horizons, they predominantly rely on historical kinematic states, treating vessel motion as an isolated system. In reality, maritime navigation is profoundly modulated by extrinsic factors like weather and constrained by static vessel characteristics. Existing multimodal approaches fundamentally model the joint distribution over states and contexts, treating environmental variables as peer features rather than encoding the directional physical dependence of vessel dynamics on environmental conditions. In this work, we propose the Conditional Informer, a novel encoder-decoder architecture that formulates trajectory prediction as a conditional generation task. We employ a dedicated Conditional Attention mechanism where the vessel state explicitly queries environmental contexts through cross-attention, encoding the physical prior that weather modulates - but is not generated by - vessel dynamics. Furthermore, to address the intermittency of real-world data, we introduce a Modality Masking training strategy to prevent catastrophic degradation during sensor fallback. Extensive experiments on AIS and ERA5 data demonstrate that our approach outperforms kinematic and concatenation-based baselines by 15.4

This paper presents a novel electronic support measures (ESM) tracking framework together with the first tracker-aware frequency scan scheduler (TAFS). Perception of the electromagnetic environment in passive radar is constrained by the receiver’s finite channel bandwidth, permitting observation of only a limited subset of frequencies at any time. Scan scheduling mitigates this limitation by dynamically selecting frequency bands to observe emitters distributed across a wide spectral range. We introduce an improved scan scheduling policy by first developing a joint angle-of-arrival (AoA)–frequency electronic support (ES) tracker that can resolve AoA ambiguities in single-platform passive radar. Next, deep reinforcement learning (RL) algorithms are trained to learn frequency scan scheduling policies that optimize tracking performance using an image-based encoding of the tracker’s multi-target belief state. This compact belief representation enables stable training and joint optimization of frequency selection and tracking accuracy. Simulation results demonstrate the feasibility and effectiveness of integrated, data-driven frequency scan scheduling for passive ES systems, improving situational awareness across wideband electromagnetic environments.

The primary purpose of the Tandem Reconnection And Cusp Electrodynamics Reconnaissance Satellites (TRACERS) Science Operations Center (SOC) is to ensure that the data necessary to achieve the TRACERS science goals are acquired, processed, and distributed to the scientific community. The SOC role in data acquisition is to facilitate science instrument planning and operations, through a weekly commanding cycle. Data processing includes generation of Level 0 and Level 1 data products, creation of Spacecraft Planet Instrument Camera-matrix Events (SPICE) kernels to provide spacecraft ephemerides and coordinate transforms for the mission, and ensuring consistency of all Level 2+ products produced by the individual instrument teams. Data distribution is undertaken in two ways. First, by hosting TRACERS data products on a public web portal during the active mission, and second by preparing mission data for transfer to the Space Physics Data Facility (SPDF) for long-term archiving.

The Analyzers for Cusp Ions (ACIs) on the TRACERS mission measure ion velocity distribution functions in the magnetospheric cusp from two closely spaced spacecraft in low Earth orbit. The precipitating and upflowing ion measurements contribute to the overarching goal of the TRACERS mission and are key to all three science objectives of the mission. ACI is a toroidal top-hat electrostatic analyzer on a spinning platform that provides full angular coverage with instantaneous 22.5° × ∼6° angular resolution for a single energy step. ACI has an ion energy range from ∼8 eV/e to 20,000 eV/e covered in 47 logarithmic-spaced energy steps with fractional energy resolution of ∼10