{"id":1093,"date":"2025-07-06T21:11:49","date_gmt":"2025-07-06T21:11:49","guid":{"rendered":"https:\/\/172-234-197-23.ip.linodeusercontent.com\/?p=1093"},"modified":"2025-07-06T21:48:10","modified_gmt":"2025-07-06T21:48:10","slug":"rf-motion-model-for-starship-landings","status":"publish","type":"post","link":"https:\/\/172-234-197-23.ip.linodeusercontent.com\/?p=1093","title":{"rendered":"RF Motion Model for Starship Landings"},"content":{"rendered":"\n<p>PODCAST: <strong>Enhanced DOMA (Dynamic Object Motion Analysis) RF Motion Model<\/strong>, implemented in Python using <strong>PyTorch<\/strong>. This model is designed to <strong>predict RF signal trajectories<\/strong>, specifically for <strong>Starship landings<\/strong>, by integrating complex environmental factors such as <strong>plasma sheath effects<\/strong> and <strong>ionospheric delays<\/strong>. The system utilizes a <strong>neural network architecture<\/strong> for predictions, incorporates <strong>GPU acceleration<\/strong> for performance, and includes functionalities for <strong>training, validation, and saving<\/strong> the model. A <strong>synthetic data generation<\/strong> component is also included to simulate realistic flight conditions and enable comprehensive testing of the model&#8217;s capabilities in predicting motion and RF signal behavior.<\/p>\n\n\n\n<figure class=\"wp-block-audio\"><audio controls src=\"http:\/\/172-234-197-23.ip.linodeusercontent.com\/wp-content\/uploads\/2025\/07\/Dynamic-Object-Motion-Analysis-for-RF-Motion-Analysis.mp3\"><\/audio><\/figure>\n\n\n\n<figure class=\"wp-block-image size-full\"><img data-opt-id=851965158  fetchpriority=\"high\" decoding=\"async\" width=\"916\" height=\"921\" src=\"https:\/\/ml6vmqguit1n.i.optimole.com\/w:auto\/h:auto\/q:mauto\/f:best\/https:\/\/172-234-197-23.ip.linodeusercontent.com\/wp-content\/uploads\/2025\/07\/image-127.png\" alt=\"\" class=\"wp-image-1101\" srcset=\"https:\/\/ml6vmqguit1n.i.optimole.com\/w:916\/h:921\/q:mauto\/f:best\/https:\/\/172-234-197-23.ip.linodeusercontent.com\/wp-content\/uploads\/2025\/07\/image-127.png 916w, https:\/\/ml6vmqguit1n.i.optimole.com\/w:298\/h:300\/q:mauto\/f:best\/https:\/\/172-234-197-23.ip.linodeusercontent.com\/wp-content\/uploads\/2025\/07\/image-127.png 298w, https:\/\/ml6vmqguit1n.i.optimole.com\/w:150\/h:150\/q:mauto\/f:best\/https:\/\/172-234-197-23.ip.linodeusercontent.com\/wp-content\/uploads\/2025\/07\/image-127.png 150w, https:\/\/ml6vmqguit1n.i.optimole.com\/w:768\/h:772\/q:mauto\/f:best\/https:\/\/172-234-197-23.ip.linodeusercontent.com\/wp-content\/uploads\/2025\/07\/image-127.png 768w\" sizes=\"(max-width: 916px) 100vw, 916px\" \/><\/figure>\n\n\n\n<p>The Enhanced DOMA (Dynamic Object Motion Analysis) RF Motion Model is specifically designed to integrate complex environmental factors to predict RF signal trajectories, particularly for Starship landings. The primary environmental factors it accounts for are <strong>plasma sheath effects<\/strong> and <strong>ionospheric effects<\/strong>.<\/p>\n\n\n\n<p><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Plasma Sheath Effects:<\/strong>\n<ul class=\"wp-block-list\">\n<li>The <code>EnhancedDOMAMotionModel<\/code> is initialized with a <code>PlasmaSheath<\/code> model. This <code>PlasmaSheath<\/code> model is configured to use a detailed model, consider magnetic effects, and account for ablation.<\/li>\n\n\n\n<li>During the prediction process, the model calculates plasma properties based on <strong>flight conditions<\/strong> such as Mach number, altitude, and velocity.<\/li>\n\n\n\n<li>From these calculations, a <code>plasma_attenuation<\/code> value is derived, specifically the attenuation in decibels for the X-band (<code>x_band<\/code> attenuation_db).<\/li>\n\n\n\n<li>This <code>plasma_attenuation<\/code> is then included as one of the input features to the neural network for predicting the next state.<\/li>\n\n\n\n<li>Furthermore, during training, a <strong>&#8220;plasma-weighted loss&#8221;<\/strong> can be applied. This means the loss function gives more emphasis (a higher weight) to training examples where plasma effects, specifically attenuation, are more significant. The weight is calculated as <code>1.0 + plasma_effects['attenuation_db'] \/ 30.0<\/code>.<\/li>\n<\/ul>\n<\/li>\n\n\n\n<li><strong>Ionospheric Effects:<\/strong>\n<ul class=\"wp-block-list\">\n<li>The <code>EnhancedDOMAMotionModel<\/code> includes an <code>ISSDataClient<\/code> to obtain ionospheric data.<\/li>\n\n\n\n<li>This client provides data such as <code>f0F2<\/code>, <code>muf<\/code>, and <code>delay<\/code>.<\/li>\n\n\n\n<li>The <code>ionospheric_delay<\/code> (measured in seconds) is extracted from this data.<\/li>\n\n\n\n<li>Similar to plasma attenuation, <code>ionospheric_delay<\/code> is also explicitly included as an input feature (one dimension) for the neural network.<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n\n\n\n<p>In summary, the model enhances its predictive capability by treating plasma attenuation and ionospheric delay as <strong>direct inputs to its neural network<\/strong>. This allows the model to learn how these complex environmental phenomena influence the RF signal trajectory (next position, rotation, and velocity) during critical phases like Starship landings. The training process can also leverage these factors through a plasma-weighted loss to improve performance in high-attenuation scenarios.<\/p>\n\n\n\n<p>Plasma effects within the Enhanced DOMA (Dynamic Object Motion Analysis) RF Motion Model are primarily calculated by a dedicated <strong><code>PlasmaSheath<\/code> model<\/strong>. This <code>PlasmaSheath<\/code> model is integrated directly into the <code>EnhancedDOMAMotionModel<\/code> upon its initialization.<\/p>\n\n\n\n<p>Here&#8217;s a breakdown of how the calculation occurs:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Model Configuration<\/strong>: The <code>PlasmaSheath<\/code> model is configured to use a <strong>detailed model<\/strong>, consider <strong>magnetic effects<\/strong>, and account for <strong>ablation<\/strong>. This indicates a comprehensive approach to modeling the plasma environment.<\/li>\n\n\n\n<li><strong>Input Parameters<\/strong>: The calculation of plasma properties depends on specific <strong>flight conditions<\/strong>. These conditions include:\n<ul class=\"wp-block-list\">\n<li><strong>Mach number<\/strong>.<\/li>\n\n\n\n<li><strong>Altitude<\/strong>.<\/li>\n\n\n\n<li><strong>Velocity<\/strong> (as a 3D numpy array).<\/li>\n<\/ul>\n<\/li>\n\n\n\n<li><strong>Calculation Method<\/strong>: The <code>PlasmaSheath<\/code> model&#8217;s <code>calculate_plasma_properties<\/code> method is invoked to perform these computations. This method takes the aforementioned flight conditions as its arguments.<\/li>\n\n\n\n<li><strong>Output<\/strong>: The primary output derived from these calculations, which is then used by the Enhanced DOMA model, is the <strong><code>plasma_attenuation<\/code><\/strong>. Specifically, the model extracts the attenuation in decibels for the <strong>X-band (<code>x_band<\/code> attenuation_db)<\/strong> from the <code>plasma_effects<\/code> dictionary returned by the <code>calculate_plasma_properties<\/code> method.<\/li>\n\n\n\n<li><strong>Integration into Prediction<\/strong>: This calculated <code>plasma_attenuation<\/code> is then used as one of the direct <strong>input features<\/strong> (a single dimension) for the neural network within the <code>EnhancedDOMAMotionModel<\/code> when predicting the next position, rotation, and velocity.<\/li>\n\n\n\n<li><strong>Training Considerations<\/strong>: During the training process, a <strong>&#8220;plasma-weighted loss&#8221;<\/strong> can be applied. This means that the loss function assigns a higher weight to training examples where plasma effects, particularly attenuation, are more significant. The weight is calculated as <code>1.0 + plasma_effects['attenuation_db'] \/ 30.0<\/code>, emphasizing scenarios with greater RF signal attenuation due to plasma. This helps the model learn to predict more accurately under challenging plasma conditions.<\/li>\n<\/ul>\n\n\n\n<p>This detailed integration ensures that the model&#8217;s predictions of RF signal trajectories are informed by and account for the dynamic and complex nature of plasma interaction during events like Starship landings.<\/p>\n\n\n\n<p>The DOMA model, specifically the <strong>Enhanced DOMA (Dynamic Object Motion Analysis) RF Motion Model<\/strong>, serves the purpose of <strong>predicting RF signal trajectories for Starship landings<\/strong>.<\/p>\n\n\n\n<p>This enhanced model is designed to integrate several complex features to achieve its goal:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>It accounts for <strong>plasma sheath effects<\/strong>.<\/li>\n\n\n\n<li>It incorporates <strong>ionospheric effects<\/strong>.<\/li>\n\n\n\n<li>It utilizes <strong>GPU acceleration<\/strong>.<\/li>\n\n\n\n<li>It ensures <strong>ML Signal Bridge compatibility<\/strong>.<\/li>\n<\/ul>\n\n\n\n<p>By taking into account factors like plasma attenuation and ionospheric delay as direct inputs to its neural network, the model learns how these environmental phenomena influence the RF signal trajectory, specifically predicting the next position, rotation, and velocity of the Starship during critical phases like landing. The training process can also use a &#8220;plasma-weighted loss&#8221; to improve performance in scenarios with significant RF signal attenuation due to plasma.<\/p>\n\n\n\n<p>The Enhanced DOMA (Dynamic Object Motion Analysis) RF Motion Model incorporates ionospheric data through a dedicated <code>ISSDataClient<\/code>.<\/p>\n\n\n\n<p>Here&#8217;s a breakdown of how this integration occurs:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Ionospheric Data Source<\/strong>: The <code>EnhancedDOMAMotionModel<\/code> is initialized with an <strong><code>ISSDataClient<\/code><\/strong>. This client is designed to obtain ionospheric data, providing mock values for <code>f0F2<\/code>, <code>muf<\/code>, and <code>delay<\/code> in MHz and seconds, respectively.<\/li>\n\n\n\n<li><strong>Data Acquisition<\/strong>: When the model performs a prediction, such as in the <code>predict_next_position<\/code> method, it calls the <code>get_ionospheric_data()<\/code> method of the <code>iss_client<\/code>. Similarly, during batch processing in <code>apply_motion_transform<\/code> or <code>generate_synthetic_data<\/code>, the <code>ISSDataClient<\/code> is used to retrieve this information.<\/li>\n\n\n\n<li><strong>Key Input Feature<\/strong>: From the data provided by the <code>ISSDataClient<\/code>, the <strong><code>ionospheric_delay<\/code><\/strong> (measured in seconds) is extracted. This <code>ionospheric_delay<\/code> is treated as a <strong>direct input feature<\/strong> for the neural network. The model&#8217;s <code>input_dim<\/code> is 7, and <code>ionospheric_delay<\/code> is one of the specific dimensions included alongside position, time, plasma attenuation, and velocity.<\/li>\n\n\n\n<li><strong>Neural Network Integration<\/strong>: The calculated <code>ionospheric_delay<\/code> is then combined with other input features (like current position, time step, plasma attenuation, and velocity) into an <code>input_data<\/code> array. This array is converted into a <code>torch.tensor<\/code> and fed into the neural network&#8217;s forward pass (<code>self(input_tensor)<\/code>) for predicting the next position, rotation, and velocity.<\/li>\n<\/ul>\n\n\n\n<p>The Enhanced DOMA (Dynamic Object Motion Analysis) RF Motion Model calculates plasma effects through a <strong>dedicated <code>PlasmaSheath<\/code> model<\/strong>. This <code>PlasmaSheath<\/code> model is a core component initialized within the <code>EnhancedDOMAMotionModel<\/code>.<\/p>\n\n\n\n<p>Here&#8217;s how the calculation process for plasma effects is incorporated:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Dedicated Plasma Model<\/strong>: The <code>EnhancedDOMAMotionModel<\/code> includes an instance of <code>PlasmaSheath<\/code>. This model is configured to be highly detailed, explicitly considering <strong>magnetic effects<\/strong> and <strong>ablation<\/strong> during its calculations.<\/li>\n\n\n\n<li><strong>Input Flight Conditions<\/strong>: To calculate plasma properties, the <code>PlasmaSheath<\/code> model requires specific <strong>flight conditions<\/strong>. These conditions are provided as a dictionary and typically include:\n<ul class=\"wp-block-list\">\n<li><strong>Mach number<\/strong> (e.g., 10.0 by default).<\/li>\n\n\n\n<li><strong>Altitude<\/strong> (e.g., 50000.0 meters by default).<\/li>\n\n\n\n<li><strong>Velocity<\/strong> (a 3D NumPy array, e.g., <code>[4000.0, 0.0, 0.0]<\/code> by default).<\/li>\n<\/ul>\n<\/li>\n\n\n\n<li><strong>Calculation Method<\/strong>: The plasma effects are computed by calling the <strong><code>calculate_plasma_properties<\/code> method<\/strong> of the <code>plasma_model<\/code> (the <code>PlasmaSheath<\/code> instance), passing in the current flight conditions.<\/li>\n\n\n\n<li><strong>Output and Integration<\/strong>: The primary output extracted from these calculations for use by the DOMA model is the <strong><code>plasma_attenuation<\/code><\/strong>. Specifically, the model retrieves the <strong>attenuation in decibels for the X-band<\/strong> (<code>x_band<\/code> <code>attenuation_db<\/code>) from the results of the plasma properties calculation. This <code>plasma_attenuation<\/code> value then becomes <strong>one of the direct input features<\/strong> for the neural network within the Enhanced DOMA model.<\/li>\n\n\n\n<li><strong>Training Adjustment<\/strong>: During the training process, the model can apply a <strong>&#8220;plasma-weighted loss&#8221;<\/strong>. This means that training examples where plasma effects, particularly attenuation, are more significant, are given a higher weight in the loss function. The weight is calculated as <code>1.0 + plasma_effects['attenuation_db'] \/ 30.0<\/code>, which helps the model to learn and perform better under challenging plasma conditions.<\/li>\n<\/ul>\n\n\n\n<p>By integrating these detailed plasma calculations, the Enhanced DOMA model ensures that its predictions of RF signal trajectories accurately reflect the impact of the dynamic plasma environment during critical phases like Starship landings.<\/p>\n\n\n\n<p>The Enhanced DOMA (Dynamic Object Motion Analysis) RF Motion Model has several key inputs that allow it to integrate complex environmental factors for RF signal trajectory prediction. The model&#8217;s neural network is configured with an <code>input_dim<\/code> of 7, which represents the combined dimensions of these various inputs.<\/p>\n\n\n\n<p>The <strong>model&#8217;s key inputs<\/strong> are:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Position (3 dimensions)<\/strong>: This refers to the current 3D position of the object, typically represented as <code>[x, y, z]<\/code> coordinates.<\/li>\n\n\n\n<li><strong>Time (1 dimension)<\/strong>: This is the current time step or time variable relevant to the prediction.<\/li>\n\n\n\n<li><strong>Plasma Attenuation (1 dimension)<\/strong>: This input specifically accounts for the <strong>attenuation in decibels for the X-band<\/strong> (<code>x_band<\/code> <code>attenuation_db<\/code>) caused by plasma sheath effects. This value is calculated by the <code>PlasmaSheath<\/code> model, which considers flight conditions such as Mach number, altitude, and velocity, and is configured to use a detailed model, magnetic effects, and ablation.<\/li>\n\n\n\n<li><strong>Ionospheric Delay (1 dimension)<\/strong>: This input represents the <code>delay<\/code> in seconds caused by ionospheric effects. This data is obtained from an <code>ISSDataClient<\/code> which provides mock values for ionospheric parameters like <code>f0F2<\/code>, <code>muf<\/code>, and <code>delay<\/code>.<\/li>\n\n\n\n<li><strong>Velocity (1 dimension)<\/strong>: This is the magnitude of the object&#8217;s velocity, derived from the 3D velocity vector provided in the flight conditions.<\/li>\n<\/ul>\n\n\n\n<p>These individual inputs are combined into a single input tensor that is fed into the neural network for predicting the next position, rotation, and velocity.<\/p>\n\n\n\n<p>Plasma effects are integrated into the Enhanced DOMA (Dynamic Object Motion Analysis) RF Motion Model through a <strong>dedicated <code>PlasmaSheath<\/code> model<\/strong> that is a core component of the <code>EnhancedDOMAMotionModel<\/code>. This integration encompasses how plasma properties are calculated, how the resulting attenuation impacts the neural network&#8217;s input, and how it influences the model&#8217;s training process.<\/p>\n\n\n\n<p>Here&#8217;s a detailed breakdown of how plasma effects are integrated:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Dedicated Plasma Calculation Model<\/strong>: The <code>EnhancedDOMAMotionModel<\/code> initializes an instance of the <code>PlasmaSheath<\/code> model. This <code>PlasmaSheath<\/code> model is configured to use a <strong>detailed model<\/strong>, consider <strong>magnetic effects<\/strong>, and account for <strong>ablation<\/strong>, indicating a comprehensive approach to simulating the plasma environment.<\/li>\n\n\n\n<li><strong>Calculation Based on Flight Conditions<\/strong>: The <code>PlasmaSheath<\/code> model calculates plasma properties based on specific <strong>flight conditions<\/strong>. These conditions typically include:\n<ul class=\"wp-block-list\">\n<li><strong>Mach number<\/strong> (e.g., <code>10.0<\/code> by default).<\/li>\n\n\n\n<li><strong>Altitude<\/strong> (e.g., <code>50000.0<\/code> meters by default).<\/li>\n\n\n\n<li><strong>Velocity<\/strong> (a 3D NumPy array, e.g., <code>[4000.0, 0.0, 0.0]<\/code> by default). The <code>calculate_plasma_properties<\/code> method of the <code>plasma_model<\/code> (the <code>PlasmaSheath<\/code> instance) is invoked with these conditions.<\/li>\n<\/ul>\n<\/li>\n\n\n\n<li><strong>Key Output: Plasma Attenuation<\/strong>: The primary output derived from these calculations, which is directly used by the DOMA model, is the <strong><code>plasma_attenuation<\/code><\/strong>. Specifically, the model extracts the <strong>attenuation in decibels for the X-band<\/strong> (<code>x_band<\/code> <code>attenuation_db<\/code>) from the <code>plasma_effects<\/code> dictionary returned by the <code>calculate_plasma_properties<\/code> method.<\/li>\n\n\n\n<li><strong>Input Feature for the Neural Network<\/strong>: This calculated <code>plasma_attenuation<\/code> is then explicitly included as <strong>one of the direct input features<\/strong> for the neural network. The model&#8217;s <code>input_dim<\/code> is 7, composed of: position (3 dimensions), time (1 dimension), plasma attenuation (1 dimension), ionospheric delay (1 dimension), and velocity (1 dimension). By feeding this value directly into the neural network, the model learns how varying plasma effects influence the RF signal trajectory, specifically predicting the next position, rotation, and velocity.<\/li>\n\n\n\n<li><strong>Training with Plasma-Weighted Loss<\/strong>: During the training process, a <strong>&#8220;plasma-weighted loss&#8221;<\/strong> can be applied. This means that the loss function assigns a higher weight to training examples where plasma effects, particularly attenuation, are more significant. The weight is calculated as <code>1.0 + plasma_effects['attenuation_db'] \/ 30.0<\/code>, which helps the model to better learn and predict in challenging scenarios where RF signal attenuation due to plasma is substantial. This mechanism biases the model to pay more attention to high-attenuation situations during learning.<\/li>\n<\/ul>\n\n\n\n<figure class=\"wp-block-image size-full\"><a href=\"https:\/\/modern-physics.org\/plasma-sheath\/\"><img data-opt-id=2046595072  fetchpriority=\"high\" decoding=\"async\" width=\"1000\" height=\"1000\" src=\"https:\/\/ml6vmqguit1n.i.optimole.com\/w:auto\/h:auto\/q:mauto\/f:best\/https:\/\/172-234-197-23.ip.linodeusercontent.com\/wp-content\/uploads\/2025\/07\/image-125.png\" alt=\"\" class=\"wp-image-1094\" srcset=\"https:\/\/ml6vmqguit1n.i.optimole.com\/w:1000\/h:1000\/q:mauto\/f:best\/https:\/\/172-234-197-23.ip.linodeusercontent.com\/wp-content\/uploads\/2025\/07\/image-125.png 1000w, https:\/\/ml6vmqguit1n.i.optimole.com\/w:300\/h:300\/q:mauto\/f:best\/https:\/\/172-234-197-23.ip.linodeusercontent.com\/wp-content\/uploads\/2025\/07\/image-125.png 300w, https:\/\/ml6vmqguit1n.i.optimole.com\/w:150\/h:150\/q:mauto\/f:best\/https:\/\/172-234-197-23.ip.linodeusercontent.com\/wp-content\/uploads\/2025\/07\/image-125.png 150w, https:\/\/ml6vmqguit1n.i.optimole.com\/w:768\/h:768\/q:mauto\/f:best\/https:\/\/172-234-197-23.ip.linodeusercontent.com\/wp-content\/uploads\/2025\/07\/image-125.png 768w\" sizes=\"(max-width: 1000px) 100vw, 1000px\" \/><\/a><\/figure>\n\n\n\n<figure class=\"wp-block-image size-large\"><img data-opt-id=1908020676  data-opt-src=\"https:\/\/ml6vmqguit1n.i.optimole.com\/w:1024\/h:776\/q:mauto\/f:best\/https:\/\/172-234-197-23.ip.linodeusercontent.com\/wp-content\/uploads\/2025\/07\/image-126.png\"  decoding=\"async\" width=\"1024\" height=\"776\" src=\"data:image/svg+xml,%3Csvg%20viewBox%3D%220%200%201024%20776%22%20width%3D%221024%22%20height%3D%22776%22%20xmlns%3D%22http%3A%2F%2Fwww.w3.org%2F2000%2Fsvg%22%3E%3Crect%20width%3D%221024%22%20height%3D%22776%22%20fill%3D%22transparent%22%2F%3E%3C%2Fsvg%3E\" alt=\"\" class=\"wp-image-1095\" old-srcset=\"https:\/\/ml6vmqguit1n.i.optimole.com\/w:1024\/h:776\/q:mauto\/f:best\/https:\/\/172-234-197-23.ip.linodeusercontent.com\/wp-content\/uploads\/2025\/07\/image-126.png 1024w, https:\/\/ml6vmqguit1n.i.optimole.com\/w:300\/h:227\/q:mauto\/f:best\/https:\/\/172-234-197-23.ip.linodeusercontent.com\/wp-content\/uploads\/2025\/07\/image-126.png 300w, https:\/\/ml6vmqguit1n.i.optimole.com\/w:768\/h:582\/q:mauto\/f:best\/https:\/\/172-234-197-23.ip.linodeusercontent.com\/wp-content\/uploads\/2025\/07\/image-126.png 768w, https:\/\/ml6vmqguit1n.i.optimole.com\/w:1042\/h:790\/q:mauto\/f:best\/https:\/\/172-234-197-23.ip.linodeusercontent.com\/wp-content\/uploads\/2025\/07\/image-126.png 1042w\" \/><\/figure>\n\n\n\n<figure class=\"wp-block-image size-full\"><img data-opt-id=1423857644  data-opt-src=\"https:\/\/ml6vmqguit1n.i.optimole.com\/w:auto\/h:auto\/q:mauto\/f:best\/https:\/\/172-234-197-23.ip.linodeusercontent.com\/wp-content\/uploads\/2025\/07\/image-129.png\"  decoding=\"async\" width=\"687\" height=\"466\" src=\"data:image/svg+xml,%3Csvg%20viewBox%3D%220%200%20100%%20100%%22%20width%3D%22100%%22%20height%3D%22100%%22%20xmlns%3D%22http%3A%2F%2Fwww.w3.org%2F2000%2Fsvg%22%3E%3Crect%20width%3D%22100%%22%20height%3D%22100%%22%20fill%3D%22transparent%22%2F%3E%3C%2Fsvg%3E\" alt=\"\" class=\"wp-image-1105\" old-srcset=\"https:\/\/ml6vmqguit1n.i.optimole.com\/w:687\/h:466\/q:mauto\/f:best\/https:\/\/172-234-197-23.ip.linodeusercontent.com\/wp-content\/uploads\/2025\/07\/image-129.png 687w, https:\/\/ml6vmqguit1n.i.optimole.com\/w:300\/h:203\/q:mauto\/f:best\/https:\/\/172-234-197-23.ip.linodeusercontent.com\/wp-content\/uploads\/2025\/07\/image-129.png 300w\" \/><\/figure>\n","protected":false},"excerpt":{"rendered":"<p>PODCAST: Enhanced DOMA (Dynamic Object Motion Analysis) RF Motion Model, implemented in Python using PyTorch. This model is designed to predict RF signal trajectories, specifically for Starship landings, by integrating complex environmental factors such as plasma sheath effects and ionospheric delays. The system utilizes a neural network architecture for predictions, incorporates GPU acceleration for performance,&hellip;&nbsp;<a href=\"https:\/\/172-234-197-23.ip.linodeusercontent.com\/?p=1093\" rel=\"bookmark\"><span class=\"screen-reader-text\">RF Motion Model for Starship Landings<\/span><\/a><\/p>\n","protected":false},"author":1,"featured_media":1104,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"neve_meta_sidebar":"","neve_meta_container":"","neve_meta_enable_content_width":"","neve_meta_content_width":0,"neve_meta_title_alignment":"","neve_meta_author_avatar":"","neve_post_elements_order":"","neve_meta_disable_header":"","neve_meta_disable_footer":"","neve_meta_disable_title":"","footnotes":""},"categories":[14,10],"tags":[],"class_list":["post-1093","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-podcast","category-signal_scythe"],"_links":{"self":[{"href":"https:\/\/172-234-197-23.ip.linodeusercontent.com\/index.php?rest_route=\/wp\/v2\/posts\/1093","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/172-234-197-23.ip.linodeusercontent.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/172-234-197-23.ip.linodeusercontent.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/172-234-197-23.ip.linodeusercontent.com\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/172-234-197-23.ip.linodeusercontent.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=1093"}],"version-history":[{"count":5,"href":"https:\/\/172-234-197-23.ip.linodeusercontent.com\/index.php?rest_route=\/wp\/v2\/posts\/1093\/revisions"}],"predecessor-version":[{"id":1106,"href":"https:\/\/172-234-197-23.ip.linodeusercontent.com\/index.php?rest_route=\/wp\/v2\/posts\/1093\/revisions\/1106"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/172-234-197-23.ip.linodeusercontent.com\/index.php?rest_route=\/wp\/v2\/media\/1104"}],"wp:attachment":[{"href":"https:\/\/172-234-197-23.ip.linodeusercontent.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=1093"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/172-234-197-23.ip.linodeusercontent.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=1093"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/172-234-197-23.ip.linodeusercontent.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=1093"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}