{"id":12027,"date":"2024-07-23T10:26:35","date_gmt":"2024-07-23T01:26:35","guid":{"rendered":"https:\/\/blogs.mathworks.com\/japan-community\/?p=12027"},"modified":"2024-08-09T23:40:29","modified_gmt":"2024-08-09T14:40:29","slug":"virtual-sensors-with-ai-and-model-based-design-jp","status":"publish","type":"post","link":"https:\/\/blogs.mathworks.com\/japan-community\/2024\/07\/23\/virtual-sensors-with-ai-and-model-based-design-jp\/","title":{"rendered":"AI \u3068 Model-Based Design \u306b\u3088\u308b\u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc"},"content":{"rendered":"<div style=\"margin: 2px 10px 9px 4px; padding: 0px; line-height: 21px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 14px; font-weight: 400; text-align: left;\">\u203b\u3053\u306e\u6295\u7a3f\u306f 2023 \u5e74 8 \u6708 11 \u65e5\u306b <a href=\"https:\/\/blogs.mathworks.com\/deep-learning\/2023\/08\/11\/virtual-sensors-with-ai-and-model-based-design\/?from=jp\">Artificial Intelligence \u3078 \u6295\u7a3f<\/a>\u3055\u308c\u305f\u3082\u306e\u306e\u6284\u8a33\u3067\u3059\u3002<\/div>\n<hr \/>\n<p>&nbsp;<\/p>\n<p>\u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc\u306f\u3001\u30ea\u30a2\u30eb\u30bf\u30a4\u30e0\u306e\u76e3\u8996\u3068\u5236\u5fa1\u304c\u5fc5\u8981\u306a\u30b7\u30b9\u30c6\u30e0\u3067\u3001\u7269\u7406\u7684\u306a\u30bb\u30f3\u30b5\u30fc\u306e\u4f7f\u7528\u304c\u5b9f\u7528\u7684\u3067\u306a\u3044\u3001\u307e\u305f\u306f\u9ad8\u4fa1\u3067\u3042\u308b\u5834\u5408\u306b\u4f7f\u7528\u3067\u304d\u307e\u3059\u3002\u305d\u3057\u3066 AI \u3092\u6d3b\u7528\u3059\u308b\u3053\u3068\u3067\u3001\u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc\u306e\u7cbe\u5ea6\u3068\u4fe1\u983c\u6027\u306e\u5411\u4e0a\u304c\u671f\u5f85\u3067\u304d\u307e\u3059\u3002\u3053\u306e\u30d6\u30ed\u30b0\u8a18\u4e8b\u3067\u306f\u3001MATLAB \u3068 Simulink \u3092\u4f7f\u7528\u3057\u3066 AI \u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc\u3092\u8a2d\u8a08\u3059\u308b\u65b9\u6cd5\u3092\u7d39\u4ecb\u3057\u307e\u3059\u3002<\/p>\n<p>\u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc\u81ea\u4f53\u306f\u65b0\u3057\u3044\u3082\u306e\u3067\u306f\u3042\u308a\u307e\u305b\u3093\u3002\u3053\u308c\u3089\u306f\u3057\u3070\u3057\u3070\u300c\u30aa\u30d6\u30b6\u30fc\u30d0\u30fc\u300d\u3068\u547c\u3070\u308c\u3001\u30a8\u30f3\u30b8\u30cb\u30a2\u306f<a href=\"https:\/\/www.mathworks.com\/videos\/series\/understanding-kalman-filters.html\">\u30ab\u30eb\u30de\u30f3\u30d5\u30a3\u30eb\u30bf<\/a>\u3084\u305d\u306e\u4ed6\u306e\u6280\u8853\uff08\u4f8b\u3048\u3070\u3001\u30d6\u30e9\u30b7\u30ec\u30b9\u30e2\u30fc\u30bf\u30fc\u30b7\u30e3\u30d5\u30c8\u306e\u4f4d\u7f6e\u3068\u901f\u5ea6\u3092\u63a8\u5b9a\u3059\u308b\u305f\u3081\u306e<a href=\"https:\/\/www.mathworks.com\/help\/mcb\/ref\/slidingmodeobserver.html\">\u30b9\u30e9\u30a4\u30c7\u30a3\u30f3\u30b0\u30e2\u30fc\u30c9\u30aa\u30d6\u30b6\u30fc\u30d0\u30fc<\/a>\u306a\u3069\uff09\u3092\u4f7f\u3063\u3066\u8a2d\u8a08\u3057\u3066\u304d\u307e\u3057\u305f\u3002AI \u306f\u5834\u5408\u306b\u3088\u3063\u3066\u306f\u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc\u306e\u8a2d\u8a08\u3092\u6539\u5584\u3067\u304d\u308b\u65b0\u3057\u3044\u9078\u629e\u80a2\u3067\u3059\u3002MATLAB \u3068 Simulink \u3092\u4f7f\u7528\u3059\u308b\u3053\u3068\u3067\u3001\u69d8\u3005\u306a\u65b9\u6cd5\u3092\u7c21\u5358\u306b\u6bd4\u8f03\u30fb\u7d44\u307f\u5408\u308f\u305b\u3066\u30ef\u30fc\u30af\u30d5\u30ed\u30fc\u306b\u6700\u9069\u306a\u89e3\u6c7a\u7b56\u3092\u898b\u3064\u3051\u308b\u3053\u3068\u304c\u3067\u304d\u307e\u3059\u3002<\/p>\n<p><img decoding=\"async\" loading=\"lazy\" class=\"alignnone wp-image-13025 size-full\" src=\"https:\/\/blogs.mathworks.com\/deep-learning\/files\/2023\/08\/virtual_sensor_ai_workflow.png\" alt=\"The workflow to design virtual sensors using AI includes data preparation, AI modeling, system simulation, and deployment\" width=\"725\" height=\"185\" \/><\/p>\n<p><strong>Figure:<\/strong> AI\u3092\u4f7f\u7528\u3057\u305f\u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc\u8a2d\u8a08\u306e\u30ef\u30fc\u30af\u30d5\u30ed\u30fc<\/p>\n<p>&nbsp;<\/p>\n<p style=\"font-size: 18px;\"><strong>\u306a\u305c\u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc\u306a\u306e\u304b\uff1f<\/strong><\/p>\n<p>\u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc\u306f\u3001\u5bfe\u8c61\u3068\u306a\u308b\u91cf\u3092\u63a8\u5b9a\u3059\u308b\u305f\u3081\u306b\u4ed6\u306e\u6e2c\u5b9a\u5024\u3084\u30c7\u30fc\u30bf\u3092\u4f7f\u7528\u3057\u3066\u3001\u7269\u7406\u7684\u306a\u30bb\u30f3\u30b5\u30fc\u306e\u52d5\u4f5c\u3092\u6a21\u5023\u3059\u308b\u30bd\u30d5\u30c8\u30a6\u30a7\u30a2\u30b3\u30f3\u30dd\u30fc\u30cd\u30f3\u30c8\u3067\u3059\u3002\u3053\u308c\u3089\u306f\u591a\u304f\u306e\u7523\u696d\u3084\u30a2\u30d7\u30ea\u30b1\u30fc\u30b7\u30e7\u30f3\u3067\u3001\u30d7\u30ed\u30bb\u30b9\u306e\u76e3\u8996\u3084\u6700\u9069\u5316\u3001\u74b0\u5883\u6761\u4ef6\u306e\u63a8\u5b9a\u306b\u5e83\u304f\u4f7f\u7528\u3055\u308c\u3066\u3044\u307e\u3059\u3002<\/p>\n<h6><\/h6>\n<h6><img decoding=\"async\" loading=\"lazy\" class=\"alignnone wp-image-13031 size-full\" src=\"https:\/\/blogs.mathworks.com\/deep-learning\/files\/2023\/08\/virtual_sensors_applications.png\" alt=\"Visual sensors can be used in many industries and applications\" width=\"1042\" height=\"607\" \/><\/h6>\n<p><strong>Figure:<\/strong> \u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc\u304c\u4f7f\u7528\u3055\u308c\u308b\u7523\u696d\u3068\u30a2\u30d7\u30ea\u30b1\u30fc\u30b7\u30e7\u30f3<\/p>\n<h6><\/h6>\n<p>\u7269\u7406\u7684\u306a\u30bb\u30f3\u30b5\u30fc\u3068\u6bd4\u8f03\u3057\u305f\u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc\u306e\u4e3b\u306a\u5229\u70b9\u306f\u3001\u30d1\u30d5\u30a9\u30fc\u30de\u30f3\u30b9\u306e\u5411\u4e0a\u3068\u30b3\u30b9\u30c8\u306e\u524a\u6e1b\u3067\u3059\u3002\u6b21\u306e\u3088\u3046\u306a\u5834\u5408\u306b\u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc\u306e\u4f7f\u7528\u3092\u691c\u8a0e\u3057\u3066\u304f\u3060\u3055\u3044\uff1a<\/p>\n<ul>\n<li>\u7269\u7406\u7684\u306a\u30bb\u30f3\u30b5\u30fc\u306e\u4f7f\u7528\u304c\u56f0\u96e3\u307e\u305f\u306f\u5b9f\u7528\u7684\u3067\u306a\u3044\u5834\u5408\u3002<\/li>\n<li>\u76f4\u63a5\u6e2c\u5b9a\u3067\u304d\u306a\u3044\u8ffd\u52a0\u60c5\u5831\u304c\u5fc5\u8981\u306a\u5834\u5408\u3002<\/li>\n<li>\u7269\u7406\u7684\u306a\u30bb\u30f3\u30b5\u30fc\u304c\u4fe1\u983c\u6027\u306b\u6b20\u3051\u308b\u3001\u7dad\u6301\u304c\u56f0\u96e3\u3001\u307e\u305f\u306f\u9ad8\u30b3\u30b9\u30c8\u3067\u3042\u308b\u5834\u5408\u3002<\/li>\n<\/ul>\n<p>&nbsp;<\/p>\n<p><strong>\u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc\u306e\u5b9f\u969b\u306e\u5fdc\u7528\u4f8b<\/strong><\/p>\n<p>MATLAB \u3068 Simulink \u306e\u30e6\u30fc\u30b6\u30fc\u306f\u3001AI\u3001\u5f93\u6765\u306e\u65b9\u6cd5\uff08\u4f8b\uff1a\u30ab\u30eb\u30de\u30f3\u30d5\u30a3\u30eb\u30bf\uff09\u3001\u304a\u3088\u3073\u305d\u308c\u3089\u3092\u7d44\u307f\u5408\u308f\u305b\u305f\u65b9\u6cd5\u7b49\u3092\u6d3b\u7528\u3057\u3066\u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc\u3092\u958b\u767a\u3057\u3066\u3044\u307e\u3059\u3002\u5f7c\u3089\u306e\u4e8b\u4f8b\u3092\u3044\u304f\u3064\u304b\u3054\u89a7\u304f\u3060\u3055\u3044\uff1a<\/p>\n<ul>\n<li><a href=\"https:\/\/jp.mathworks.com\/company\/user_stories\/coca-cola-develops-virtual-pressure-sensor-with-machine-learning-to-improve-beverage-dispenser-diagnostics.html\">Coca-Cola\u3001\u6a5f\u68b0\u5b66\u7fd2\u3092\u4f7f\u7528\u3057\u305f\u4eee\u60f3\u5727\u529b\u30bb\u30f3\u30b5\u30fc\u306e\u958b\u767a\u306b\u3088\u308a\u98f2\u6599\u30c7\u30a3\u30b9\u30da\u30f3\u30b5\u30fc\u306e\u8a3a\u65ad\u3092\u6539\u5584<br \/>\n<\/a><\/li>\n<li><a href=\"https:\/\/jp.mathworks.com\/company\/user_stories\/mercedes-benz-simulates-hardware-sensors-with-deep-neural-networks.html?s_tid=srchtitle_virtual%20sensor_1\">Mercedes-Benz\u3001\u30c7\u30a3\u30fc\u30d7 \u30cb\u30e5\u30fc\u30e9\u30eb \u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u3067\u30cf\u30fc\u30c9\u30a6\u30a7\u30a2\u30bb\u30f3\u30b5\u30fc\u3092\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3<\/a><\/li>\n<li><a href=\"https:\/\/www.mathworks.com\/company\/user_stories\/poclain-hydraulics-develops-soft-sensors-to-measure-motor-remperature-in-real-time-using-deep-learning-and-kalman-filters.html\">Poclain Hydraulics Develops Soft Sensors to Measure Motor Temperature in Real Time Using Deep Learning and Kalman Filters<\/a><\/li>\n<li><a href=\"https:\/\/www.mathworks.com\/videos\/developing-onboard-soh-estimation-using-dva-and-ica-for-lfp-batteries-1654073089339.html\">Gotion Develops Onboard SOH Estimation Using DVA and ICA for LFP Batteries<\/a><\/li>\n<li><a href=\"https:\/\/www.mathworks.com\/content\/dam\/mathworks\/mathworks-dot-com\/company\/events\/conferences\/automotive-conference-michigan\/2022\/battery-soh-and-soc-estimation-using-a-hybrid-machine-learning-approach.pdf\" target=\"_blank\">Battery SOC and SOH Estimation using a Hybrid Machine Learning Approach<\/a><\/li>\n<li><a href=\"https:\/\/www.mathworks.com\/content\/dam\/mathworks\/mathworks-dot-com\/company\/events\/conferences\/automotive-conference-michigan\/2022\/onboard-battery-pack-state-of-charge-estimation-using-a-trained-neural-network.pdf\" target=\"_blank\">Onboard Battery Pack State of Charge Estimation Using a Neural Network<\/a><\/li>\n<\/ul>\n<p>&nbsp;<\/p>\n<p style=\"font-size: 18px;\"><strong>AI\u3092\u7528\u3044\u305f\u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc<\/strong><\/p>\n<p>\u3053\u306e\u30bb\u30af\u30b7\u30e7\u30f3\u3067\u306f\u3001AI\u3092\u4f7f\u7528\u3057\u3066\u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc\u3092\u8a2d\u8a08\u3059\u308b\u65b9\u6cd5\u306b\u3064\u3044\u3066\u8aac\u660e\u3057\u3001\u4ee5\u4e0b\u306e\u70b9\u3092\u793a\u3057\u307e\u3059\uff1a<\/p>\n<ul>\n<li>MATLAB \u3067\u3055\u307e\u3056\u307e\u306a AI \u30e2\u30c7\u30eb\u3092\u8fc5\u901f\u306b\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u304a\u3088\u3073\u6bd4\u8f03\u3057\u3066\u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc\u3092\u5b9f\u88c5\u3059\u308b\u3001\u307e\u305f\u306f\u4ed6\u306e\u30c7\u30a3\u30fc\u30d7\u30e9\u30fc\u30cb\u30f3\u30b0\u30d5\u30ec\u30fc\u30e0\u30ef\u30fc\u30af\u304b\u3089 AI \u30e2\u30c7\u30eb\u3092\u30a4\u30f3\u30dd\u30fc\u30c8\u3059\u308b\u3053\u3068\u304c\u3067\u304d\u307e\u3059\u3002<\/li>\n<li>\u672c\u756a\u74b0\u5883\u306b\u5c55\u958b\u3059\u308b\u524d\u306b\u3001\u3055\u307e\u3056\u307e\u306a\u30c6\u30b9\u30c8\u30b7\u30ca\u30ea\u30aa\u3092\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u3057\u3066\u3001\u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc\u306e\u8a2d\u8a08\u3092\u5b89\u5168\u306b\u3001\u4f53\u7cfb\u7684\u306b\u3001\u5fb9\u5e95\u7684\u306b\u30c6\u30b9\u30c8\u3059\u308b\u3053\u3068\u304c\u3067\u304d\u307e\u3059\u3002<\/li>\n<\/ul>\n<p>\u30d0\u30c3\u30c6\u30ea\u30fc\u306e\u5145\u96fb\u72b6\u614b\uff08SOC\uff09\u63a8\u5b9a\u306e\u305f\u3081\u306e AI \u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc\u306e\u4f8b\u3092\u4f7f\u7528\u3057\u3066\u3001\u8a2d\u8a08\u624b\u9806\u3092\u8aac\u660e\u3057\u3001AI \u3092\u4f7f\u7528\u3059\u308b\u5229\u70b9\u3092\u5f37\u8abf\u3057\u307e\u3059\u3002\u4f8b\u306e\u8a73\u7d30\u306f<a href=\"https:\/\/www.mathworks.com\/campaigns\/offers\/next\/integrating-ai-into-system-level-design\/using-ai-for-embedded-algorithms.html\">\u3053\u3061\u3089<\/a>\u3067\u78ba\u8a8d\u3067\u304d\u307e\u3059\u3002SOC \u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc\u306f\u3001\u30d0\u30c3\u30c6\u30ea\u30fc\u30de\u30cd\u30b8\u30e1\u30f3\u30c8\u30b7\u30b9\u30c6\u30e0\uff08BMS\uff09\u306e\u91cd\u8981\u306a\u30b3\u30f3\u30dd\u30fc\u30cd\u30f3\u30c8\u3067\u3042\u308a\u3001\u30d0\u30c3\u30c6\u30ea\u30fc\u306e\u5b89\u5168\u304b\u3064\u52b9\u7387\u7684\u306a\u904b\u7528\u3092\u4fdd\u8a3c\u3057\u307e\u3059\u3002\u30ea\u30c1\u30a6\u30e0\u30a4\u30aa\u30f3\u96fb\u6c60\u306f\u3001\u30a6\u30a7\u30a2\u30e9\u30d6\u30eb\u96fb\u5b50\u6a5f\u5668\u3001\u643a\u5e2f\u96fb\u8a71\u3001\u30ce\u30fc\u30c8\u30d1\u30bd\u30b3\u30f3\u304b\u3089\u96fb\u6c17\u81ea\u52d5\u8eca\u3001\u30b9\u30de\u30fc\u30c8\u30b0\u30ea\u30c3\u30c9\u306b\u81f3\u308b\u307e\u3067\u3001\u4eca\u65e5\u3042\u3089\u3086\u308b\u5834\u6240\u3067\u4f7f\u7528\u3055\u308c\u3066\u3044\u307e\u3059\u3002<\/p>\n<p><img decoding=\"async\" loading=\"lazy\" class=\"alignnone wp-image-13040 size-full\" src=\"https:\/\/blogs.mathworks.com\/deep-learning\/files\/2023\/08\/SoC_video.gif\" alt=\"Measurements of voltage, current, and temperature are used to train AI model that estimates State of Charge\" width=\"1414\" height=\"370\" \/><\/p>\n<p><strong>Figure:<\/strong> \u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc\u306b\u3088\u308b\u30d0\u30c3\u30c6\u30ea\u30fc\u306e\u5145\u96fb\u72b6\u614b (SOC) \u63a8\u5b9a<\/p>\n<h6><\/h6>\n<p>\u3053\u306e\u30d6\u30ed\u30b0\u8a18\u4e8b\u3067\u306f SOC \u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc\u306e\u4f8b\u3092\u901a\u3058\u3066\u8aac\u660e\u3057\u3066\u3044\u307e\u3059\u304c\u3001SOC \u306e\u4f8b\u306b\u95a2\u9023\u3057\u306a\u3044\u30c4\u30fc\u30eb\u3084\u30aa\u30d7\u30b7\u30e7\u30f3\u306b\u3064\u3044\u3066\u3082\u89e6\u308c\u3066\u3044\u307e\u3059\u3002\u3053\u308c\u3089\u306e\u30c4\u30fc\u30eb\u3084\u30aa\u30d7\u30b7\u30e7\u30f3\u306f\u3001\u3042\u306a\u305f\u306e\u30ef\u30fc\u30af\u30d5\u30ed\u30fc\u306b\u5f79\u7acb\u3064\u304b\u3082\u3057\u308c\u307e\u305b\u3093\u3002<\/p>\n<h6><\/h6>\n<p><strong>\u30c7\u30fc\u30bf\u306e\u6e96\u5099<\/strong><\/p>\n<p>\u8a2d\u8a08\u30d7\u30ed\u30bb\u30b9\u306f\u3001\u30c7\u30fc\u30bf\u306e\u54c1\u8cea\u3084\u5229\u7528\u53ef\u80fd\u6027\u306b\u3088\u3063\u3066\u5236\u7d04\u3055\u308c\u308b\u3053\u3068\u304c\u3042\u308a\u307e\u3059\u3002MATLAB \u306f\u3001\u6b21\u306e\u65b9\u6cd5\u3067\u30c7\u30fc\u30bf\u306e\u54c1\u8cea\u3092\u5411\u4e0a\u3055\u305b\u308b\u306e\u306b\u5f79\u7acb\u3061\u307e\u3059\uff1a(1) <a href=\"https:\/\/www.mathworks.com\/solutions\/artificial-intelligence\/ai-signal-processing.html\">\u4fe1\u53f7\u51e6\u7406\u6280\u8853<\/a>\u3092\u4f7f\u7528\u3057\u3066\u30ce\u30a4\u30ba\u3092\u9664\u53bb\u3059\u308b\u3001(2) \u30c7\u30fc\u30bf\u304b\u3089\u610f\u5473\u306e\u3042\u308b\u6a19\u6e96\u7684\u304a\u3088\u3073\u30c9\u30e1\u30a4\u30f3\u56fa\u6709\u306e<a href=\"https:\/\/www.mathworks.com\/discovery\/feature-extraction.html\">\u7279\u5fb4\u91cf<\/a>\u3092\u62bd\u51fa\u3059\u308b\u3002\u307e\u305f\u3001\u30ed\u30fc\u30b3\u30fc\u30c9\u306e\u30a2\u30d7\u30ea\uff08\u4f8b\uff1a<a href=\"https:\/\/www.mathworks.com\/help\/signal\/ref\/signalanalyzer-app.html\">Signal Analyzer<\/a>\uff09\u3092\u4f7f\u7528\u3057\u3066\u30c7\u30fc\u30bf\u3092\u63a2\u7d22\u3059\u308b\u3053\u3068\u3082\u3067\u304d\u307e\u3059\u3002<\/p>\n<p>\u7279\u306b\u30c7\u30a3\u30fc\u30d7\u30e9\u30fc\u30cb\u30f3\u30b0\u30e2\u30c7\u30eb\u306f\u3001\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u306e\u305f\u3081\u306b\u5927\u898f\u6a21\u306a\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u3092\u5fc5\u8981\u3068\u3057\u307e\u3059\u3002\u30c7\u30fc\u30bf\u306e\u53ef\u7528\u6027\u306e\u554f\u984c\u3092\u514b\u670d\u3059\u308b\u305f\u3081\u306b\u3001\u5b9f\u4e16\u754c\u306e\u30c7\u30fc\u30bf\u3092\u88dc\u5b8c\u3059\u308b\u305f\u3081\u306b\u5408\u6210\u30c7\u30fc\u30bf\u3092\u4f7f\u7528\u3059\u308b\u3053\u3068\u304c\u3067\u304d\u307e\u3059\u3002\u751f\u6210\u7684\u6575\u5bfe\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\uff08<a href=\"https:\/\/www.mathworks.com\/discovery\/generative-adversarial-networks.html\">GAN<\/a>\uff09\u3092\u4f7f\u7528\u3057\u3066\u5408\u6210\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u3092\u751f\u6210\u3057\u305f\u308a\u3001<a href=\"https:\/\/www.mathworks.com\/content\/dam\/mathworks\/white-paper\/gated\/deep-learning-for-comms-and-radar-white-paper.pdf\" target=\"_blank\">\u6ce2\u5f62\u30bf\u30a4\u30d7\u3092\u7d44\u307f\u5408\u308f\u305b<\/a>\u305f\u308a\uff08\u901a\u4fe1\u304a\u3088\u3073\u30ec\u30fc\u30c0\u30fc\u30a2\u30d7\u30ea\u30b1\u30fc\u30b7\u30e7\u30f3\u7528\uff09\u3001Simulink \u3067\u7269\u7406\u30b7\u30b9\u30c6\u30e0\u306e\u30e2\u30c7\u30eb\u3092\u4f5c\u6210\u304a\u3088\u3073\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u3059\u308b\u3053\u3068\u306b\u3088\u3063\u3066\u3001\u5408\u6210\u30c7\u30fc\u30bf\u3092\u751f\u6210\u3067\u304d\u307e\u3059\u3002<\/p>\n<h6><\/h6>\n<p><strong>AI \u30e2\u30c7\u30ea\u30f3\u30b0<\/strong><\/p>\n<p>\u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc\u3092\u958b\u767a\u3059\u308b\u305f\u3081\u306b\u3001<a href=\"https:\/\/www.mathworks.com\/discovery\/machine-learning-models.html\">\u6a5f\u68b0\u5b66\u7fd2\u30a2\u30eb\u30b4\u30ea\u30ba\u30e0<\/a>\uff08\u4f8b\uff1aSVM\u3001\u7dda\u5f62\u56de\u5e30\u3001\u30cb\u30e5\u30fc\u30e9\u30eb\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\uff09\u3092\u4f7f\u7528\u3057\u3066\u6570\u884c\u306e MATLAB \u30b3\u30fc\u30c9\u3067 AI \u30e2\u30c7\u30eb\u3092\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u3059\u308b\u304b\u3001TensorFlow\u2122 \u3084 PyTorch\u00ae \u3067\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u3057\u305f<a href=\"https:\/\/www.mathworks.com\/solutions\/deep-learning\/models.html\">\u30e2\u30c7\u30eb\u3092\u30a4\u30f3\u30dd\u30fc\u30c8\u3059\u308b<\/a>\u3053\u3068\u304c\u3067\u304d\u307e\u3059\u3002<\/p>\n<p>\u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc\u306b\u5f79\u7acb\u3064\u30e2\u30c7\u30eb\u3092\u4f5c\u6210\u3059\u308b\u65b9\u6cd5\u3092\u898b\u3066\u307f\u307e\u3057\u3087\u3046<\/p>\n<p>1. sequence-to-one \u56de\u5e30\u306e\u305f\u3081\u306e<a href=\"https:\/\/www.mathworks.com\/help\/deeplearning\/ug\/long-short-term-memory-networks.html\">LSTM\uff08\u9577\u77ed\u671f\u8a18\u61b6\uff09<\/a>\u30e2\u30c7\u30eb\u3092\u4f5c\u6210\u3057\u307e\u3059\u3002LSTM \u306f\u3001\u9577\u3055\u304c\u5909\u52d5\u3059\u308b\u6642\u7cfb\u5217\u30c7\u30fc\u30bf\u3092\u6271\u3046\u306e\u306b\u7279\u306b\u52b9\u679c\u7684\u3067\u3059\u3002\u30a8\u30f3\u30b8\u30f3\u306e\u6b8b\u308a\u306e\u6709\u7528\u5bff\u547d\uff08RUL\uff09\u3092\u4e88\u6e2c\u3059\u308b\u65b9\u6cd5\u306e\u4f8b\u306b\u3064\u3044\u3066\u306f\u3001\u300c<a href=\"https:\/\/mathworks.com\/help\/deeplearning\/ug\/sequence-to-sequence-regression-using-deep-learning.html\">\u6df1\u5c64\u5b66\u7fd2\u3092\u4f7f\u7528\u3057\u305f sequence-to-sequence \u56de\u5e30<\/a>\u300d\u3092\u53c2\u7167\u3057\u3066\u304f\u3060\u3055\u3044\u3002<\/p>\n<pre>layers = [sequenceInputLayer(numFeatures)\nlstmLayer(numHiddenUnits,OutputMode=\"last\")\nfullyConnectedLayer(numResponses)\nregressionLayer];\n<\/pre>\n<p>\u305d\u3057\u3066 <code>trainNetwork<\/code>\u95a2\u6570\u3092\u4f7f\u7528\u3057\u3066LSTM\u30e2\u30c7\u30eb\u3092\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u3057\u307e\u3059\u3002<\/p>\n<h6><\/h6>\n<p>2. \u56de\u5e30\u306e\u305f\u3081\u306e\u30d5\u30a3\u30fc\u30c9\u30d5\u30a9\u30ef\u30fc\u30c9\u3001\u5168\u7d50\u5408\u30cb\u30e5\u30fc\u30e9\u30eb\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u3092\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u3057\u307e\u3059\u3002<\/p>\n<pre>Mdl = fitrnet(X,Y)\n<\/pre>\n<h6><\/h6>\n<p>3. \u56de\u5e30\u306e\u305f\u3081\u306e\u6c7a\u5b9a\u6728\u3092\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u3057\u307e\u3059\u3002<\/p>\n<pre>tree = fitrtree(X,Y);\n<\/pre>\n<p>\u3088\u308a\u30a4\u30f3\u30bf\u30e9\u30af\u30c6\u30a3\u30d6\u306a\u30ef\u30fc\u30af\u30d5\u30ed\u30fc\u3092\u597d\u3080\u5834\u5408\u306f\u3001\u30ed\u30fc\u30b3\u30fc\u30c9\u306e\u30a2\u30d7\u30ea\uff08<a href=\"https:\/\/www.mathworks.com\/help\/deeplearning\/ref\/deepnetworkdesigner-app.html\">Deep Network Designer<\/a>\u3001<a href=\"https:\/\/www.mathworks.com\/help\/stats\/classificationlearner-app.html\">Classification Learner<\/a>\u3001<a href=\"https:\/\/www.mathworks.com\/help\/stats\/regressionlearner-app.html\">Regression Learner<\/a>\uff09\u3092\u4f7f\u7528\u3057\u3066\u3001AI \u30e2\u30c7\u30eb\u306e\u8a2d\u8a08\u3001\u30c1\u30e5\u30fc\u30cb\u30f3\u30b0\u3001\u8a55\u4fa1\u3001\u304a\u3088\u3073\u6700\u9069\u5316\u3092\u884c\u3046\u3053\u3068\u304c\u3067\u304d\u307e\u3059\u3002<\/p>\n<p>MATLAB \u306f\u3001<a href=\"https:\/\/www.mathworks.com\/help\/ident\/nonlinear-model-identification.html\">\u30b7\u30b9\u30c6\u30e0\u30c0\u30a4\u30ca\u30df\u30af\u30b9\u3092\u6349\u3048\u308b\u305f\u3081\u306e\u975e\u7dda\u5f62\u30e2\u30c7\u30eb<\/a>\u3082\u63d0\u4f9b\u3057\u3066\u3044\u307e\u3059\u3002Hammerstein-Wiener \u3084 Nonlinear ARX \u30e2\u30c7\u30eb\u3092\u4f7f\u7528\u3057\u3066\u3001\u30ac\u30a6\u30b9\u904e\u7a0b\uff08GP\uff09\u3001\u30b5\u30dd\u30fc\u30c8\u30d9\u30af\u30bf\u30fc\u30de\u30b7\u30f3\uff08SVM\uff09\u3001\u305d\u306e\u4ed6\u306e\u8868\u73fe\u306a\u3069\u306e\u6a5f\u68b0\u5b66\u7fd2\u6280\u8853\u3092\u7528\u3044\u3066\u975e\u7dda\u5f62\u30b7\u30b9\u30c6\u30e0\u30c0\u30a4\u30ca\u30df\u30af\u30b9\u3092\u63a8\u5b9a\u3067\u304d\u307e\u3059\u3002\u3042\u308b\u3044\u306f\u3001\u30c7\u30a3\u30fc\u30d7\u30e9\u30fc\u30cb\u30f3\u30b0\u3092\u4f7f\u7528\u3057\u3066\u30cb\u30e5\u30fc\u30e9\u30eb\u5e38\u5fae\u5206\u65b9\u7a0b\u5f0f\uff08ODE\uff09\u30e2\u30c7\u30eb\u3092\u4f5c\u6210\u3057\u3001\u975e\u7dda\u5f62\u30b7\u30b9\u30c6\u30e0\u30c0\u30a4\u30ca\u30df\u30af\u30b9\u3092\u6349\u3048\u308b\u3053\u3068\u3082\u3067\u304d\u307e\u3059\u3002\u4f8b\u306b\u3064\u3044\u3066\u306f\u3001\u300c<a href=\"https:\/\/jp.mathworks.com\/help\/ident\/ug\/nonlinear-state-space-model-of-si-engine-torque-dynamics.html\">SI\u30a8\u30f3\u30b8\u30f3\u30c8\u30eb\u30af\u30c0\u30a4\u30ca\u30df\u30af\u30b9\u306e\u30cb\u30e5\u30fc\u30e9\u30eb\u72b6\u614b\u7a7a\u9593\u30e2\u30c7\u30eb<\/a>\u300d\u3092\u53c2\u7167\u3057\u3066\u304f\u3060\u3055\u3044\u3002<\/p>\n<p>AI \u30e2\u30c7\u30eb\u306f\u3001\u30ab\u30eb\u30de\u30f3\u30d5\u30a3\u30eb\u30bf\u306a\u3069\u306e\u4ed6\u306e\u65b9\u6cd5\u3068\u7d44\u307f\u5408\u308f\u305b\u3066\u4f7f\u7528\u3059\u308b\u3053\u3068\u304c\u3067\u304d\u307e\u3059\u3002\u305f\u3068\u3048\u3070\u3001\u30cb\u30e5\u30fc\u30e9\u30eb\u72b6\u614b\u7a7a\u9593\u30e2\u30c7\u30eb\u3092\u975e\u7dda\u5f62\u30ab\u30fc\u30de\u30f3\u30d5\u30a3\u30eb\u30bf\u306e\u5185\u90e8\u30e2\u30c7\u30eb\u3068\u3057\u3066\u4f7f\u7528\u3059\u308b\u3053\u3068\u304c\u3067\u304d\u307e\u3059\uff08\u4f8b\uff1a\u300c<a href=\"https:\/\/www.mathworks.com\/help\/ident\/ug\/online-state-estimation-using-identified-models-nonlinear-models.html\">\u540c\u5b9a\u30e2\u30c7\u30eb\u3092\u4f7f\u7528\u3057\u305f\u30aa\u30f3\u30e9\u30a4\u30f3\u72b6\u614b\u63a8\u5b9a &#8211; \u975e\u7dda\u5f62\u30e2\u30c7\u30eb<\/a>\u300d\uff09\u3002<\/p>\n<h6><\/h6>\n<p><strong>\u30b7\u30b9\u30c6\u30e0\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3<\/strong><\/p>\n<p>AI \u30e2\u30c7\u30eb\u3092\u4f5c\u6210\u307e\u305f\u306f\u8aad\u307f\u8fbc\u3093\u3060\u3089\u3001\u305d\u308c\u3089\u3092 Simulink \u306b\u7d71\u5408\u3057\u3066\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u3067\u30c6\u30b9\u30c8\u3059\u308b\u3053\u3068\u304c\u3067\u304d\u307e\u3059\u3002Simulink \u3092\u4f7f\u7528\u3059\u308b\u3068\u3001\u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc\u304c\u5c5e\u3059\u308b\u5927\u898f\u6a21\u306a\u30b7\u30b9\u30c6\u30e0\uff08\u4f8b\u3068\u3057\u3066 BMS\uff09\u306e\u6587\u8108\u3067\u3001\u7cbe\u5ea6\u3001\u30e2\u30c7\u30eb\u30b5\u30a4\u30ba\u3001\u63a8\u8ad6\u901f\u5ea6\u306a\u3069\u306e\u30c8\u30ec\u30fc\u30c9\u30aa\u30d5\u3092\u8a55\u4fa1\u3059\u308b\u3053\u3068\u304c\u3067\u304d\u307e\u3059\u3002SOC \u63a8\u5b9a\u306e\u4f8b\u3067\u306f\u3001Simulink \u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u3067 4 \u3064\u306e\u30e2\u30c7\u30eb\u304c\u6bd4\u8f03\u3055\u308c\u307e\u3059\uff1a\u62e1\u5f35\u30ab\u30eb\u30de\u30f3\u30d5\u30a3\u30eb\u30bf\uff08EKF\uff09\u3001\u56de\u5e30\u6728\u3001\u30d5\u30a3\u30fc\u30c9\u30d5\u30a9\u30ef\u30fc\u30c9\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\uff08DL-FFN\uff09\u3001\u304a\u3088\u3073LSTM\u30e2\u30c7\u30eb\u3002<\/p>\n<p><img decoding=\"async\" loading=\"lazy\" class=\"alignnone wp-image-13046 size-full\" src=\"https:\/\/blogs.mathworks.com\/deep-learning\/files\/2023\/08\/system_simulation.jpg\" alt=\"Simulink simulation of an extended Kalman filter (EKF), a regression tree, a feed-forward network (DL-FFN), and an LSTM model. \" width=\"1200\" height=\"890\" \/><\/p>\n<p><strong>Figure:<\/strong> Simulink\u30674\u3064\u306e\u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc\uff08\u30ab\u30eb\u30de\u30f3\u30d5\u30a3\u30eb\u30bf\u30e2\u30c7\u30eb\u30683\u3064\u306eAI\u30e2\u30c7\u30eb\uff09\u3092\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u304a\u3088\u3073\u6bd4\u8f03\u3059\u308b<\/p>\n<p>&nbsp;<\/p>\n<p>AI \u30e2\u30c7\u30eb\u306b\u57fa\u3065\u304f\u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc\u3092\u4f7f\u7528\u3059\u308b\u4e3b\u306a\u5229\u70b9\u306e\u4e00\u3064\u306f\u3001\u7269\u7406\u30e2\u30c7\u30eb\uff08\u7279\u306b\u8907\u96d1\u306a\u7269\u7406\u30e2\u30c7\u30eb\uff09\u306b\u57fa\u3065\u304f\u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc\u3088\u308a\u3082\u9ad8\u901f\u3067\u3042\u308b\u53ef\u80fd\u6027\u304c\u3042\u308b\u3053\u3068\u3067\u3059\u3002\u4e00\u65b9\u3067\u3001AI \u30e2\u30c7\u30eb\u306f\u30ab\u30eb\u30de\u30f3\u30d5\u30a3\u30eb\u30bf\u30e2\u30c7\u30eb\u306b\u6bd4\u3079\u3066<a href=\"https:\/\/blogs.mathworks.com\/deep-learning\/2022\/12\/30\/what-is-explainable-ai\/\">\u8aac\u660e\u6027<\/a>\u304c\u4f4e\u304f\u3001AI \u30e2\u30c7\u30eb\u306e\u5185\u90e8\u52d5\u4f5c\u3092\u7406\u89e3\u3059\u308b\u305f\u3081\u306b\u306f<a href=\"https:\/\/www.mathworks.com\/discovery\/interpretability.html\">\u8ffd\u52a0\u306e\u6280\u8853<\/a>\u304c\u5fc5\u8981\u306b\u306a\u308a\u307e\u3059\u3002\u3057\u304b\u3057\u3001\u8aac\u660e\u6027\u306e\u6b20\u5982\u306f\u5fc5\u305a\u3057\u3082\u554f\u984c\u3067\u306f\u306a\u3044\u3068\uff08<a href=\"https:\/\/www.mathworks.com\/company\/user_stories\/poclain-hydraulics-develops-soft-sensors-to-measure-motor-remperature-in-real-time-using-deep-learning-and-kalman-filters.html\">Poclain Hydraulics\u306e\u30a8\u30f3\u30b8\u30cb\u30a2\u306b\u3088\u308c\u3070<\/a>\uff09\u3055\u308c\u3066\u3044\u307e\u3059\u3002<\/p>\n<p>\u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc\u3092\u4f7f\u7528\u3057\u3066\u5b8c\u5168\u306a\u30b7\u30b9\u30c6\u30e0\u3092\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u304a\u3088\u3073\u30c6\u30b9\u30c8\u3059\u308b\u3053\u3068\u3067\u3001\u901f\u5ea6\u3084\u7cbe\u5ea6\u306a\u3069\u306e\u30e2\u30c7\u30eb\u30d1\u30d5\u30a9\u30fc\u30de\u30f3\u30b9\u3092\u8a55\u4fa1\u3057\u3001\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u306e\u7d50\u679c\u3092\u57fa\u306b\u30e2\u30c7\u30eb\u9078\u629e\u306e\u5224\u65ad\u3092\u884c\u3046\u3053\u3068\u304c\u3067\u304d\u307e\u3059\u3002<\/p>\n<p>&nbsp;<\/p>\n<p><strong>\u30c7\u30d7\u30ed\u30a4\u30e1\u30f3\u30c8<\/strong><\/p>\n<p>MATLAB \u304a\u3088\u3073\/\u307e\u305f\u306f Simulink \u3092\u4f7f\u7528\u3057\u3066\u3001\u30ea\u30bd\u30fc\u30b9\u5236\u7d04\u306e\u3042\u308b<a href=\"https:\/\/www.mathworks.com\/solutions\/deep-learning\/edge-ai.html\">\u30a8\u30c3\u30b8\u30b7\u30b9\u30c6\u30e0<\/a>\u306b AI \u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc\u3092\u30c7\u30d7\u30ed\u30a4\u3059\u308b\u305f\u3081\u306e\u30e9\u30a4\u30d6\u30e9\u30ea\u30d5\u30ea\u30fc\u306e C\/C++ \u30b3\u30fc\u30c9\u3092\u751f\u6210\u3067\u304d\u307e\u3059\u3002\u307e\u305f\u3001FPGA \u30c7\u30d0\u30a4\u30b9\u3001\u30a8\u30f3\u30bf\u30fc\u30d7\u30e9\u30a4\u30ba\u30b7\u30b9\u30c6\u30e0\u3001\u307e\u305f\u306f<a href=\"https:\/\/www.mathworks.com\/help\/deeplearning\/ug\/deep-learning-in-the-cloud.html\">\u30af\u30e9\u30a6\u30c9<\/a>\u306b\u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc\u3092\u30c7\u30d7\u30ed\u30a4\u3059\u308b\u3053\u3068\u3082\u53ef\u80fd\u3067\u3059\u3002<\/p>\n<p>\u3053\u306e\u6bb5\u968e\u3067\u306f\u3001\u30d7\u30ed\u30bb\u30c3\u30b5\u30a4\u30f3\u30b6\u30eb\u30fc\u30d7\uff08<a href=\"https:\/\/www.mathworks.com\/help\/ecoder\/ug\/about-sil-and-pil-simulations.html\">PIL<\/a>\uff09\u307e\u305f\u306f\u30cf\u30fc\u30c9\u30a6\u30a7\u30a2\u30a4\u30f3\u30b6\u30eb\u30fc\u30d7\uff08<a href=\"https:\/\/www.mathworks.com\/discovery\/hardware-in-the-loop-hil.html\">HIL<\/a>\uff09\u30c6\u30b9\u30c8\u3092\u5b9f\u884c\u3067\u304d\u307e\u3059\u3002\u30d0\u30c3\u30c6\u30ea\u30fc SOC \u3092\u63a8\u5b9a\u3059\u308b\u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc\u3092\u542b\u3080 BMS \u30bd\u30d5\u30c8\u30a6\u30a7\u30a2\u5168\u4f53\u3092\u30bf\u30fc\u30b2\u30c3\u30c8\u30cf\u30fc\u30c9\u30a6\u30a7\u30a2\u306b\u30c7\u30d7\u30ed\u30a4\u3057\u3001\u30d7\u30e9\u30f3\u30c8\u30e2\u30c7\u30eb\uff08\u3053\u306e\u5834\u5408\u306f\u30d0\u30c3\u30c6\u30ea\u30fc\u30c0\u30a4\u30ca\u30df\u30af\u30b9\u306e\u30e2\u30c7\u30eb\uff09\u3068\u30ea\u30a2\u30eb\u30bf\u30a4\u30e0\u306e\u9589\u30eb\u30fc\u30d7\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u3092\u5b9f\u884c\u3067\u304d\u307e\u3059\u3002\u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc\u306e\u4e3b\u8981\u8a2d\u8a08\u30d1\u30e9\u30e1\u30fc\u30bf\uff08\u7cbe\u5ea6\u3001\u30e1\u30e2\u30ea\u4f7f\u7528\u91cf\u3001\u63a8\u8ad6\u901f\u5ea6\u306a\u3069\uff09\u3092\u30c6\u30b9\u30c8\u3059\u308b\u3053\u3068\u304c\u3067\u304d\u307e\u3059\u3002\u30c6\u30b9\u30c8\u304c\u5b8c\u4e86\u3059\u308b\u3068\u3001\u30e2\u30c7\u30eb\u306f\u672c\u756a\u30cf\u30fc\u30c9\u30a6\u30a7\u30a2\u306b\u30c7\u30d7\u30ed\u30a4\u3059\u308b\u6e96\u5099\u304c\u6574\u3044\u307e\u3059\u3002<\/p>\n<h6><img decoding=\"async\" loading=\"lazy\" class=\"alignnone wp-image-13049 size-full\" src=\"https:\/\/blogs.mathworks.com\/deep-learning\/files\/2023\/08\/SOC_deployment.jpg\" alt=\"Deployment of SoC virtual sensor to production hardware\" width=\"1080\" height=\"195\" \/><\/h6>\n<p><strong>Figure: <\/strong>AI\u30e2\u30c7\u30eb\u306eC\/C++\u30b3\u30fc\u30c9\u3092\u751f\u6210\u3057\u3001\u8eca\u4e21\u3067\u4f7f\u7528\u3059\u308b\u30cf\u30fc\u30c9\u30a6\u30a7\u30a2\u306b\u30c7\u30d7\u30ed\u30a4\u3059\u308b\u3002<\/p>\n<p>&nbsp;<\/p>\n<p>MATLAB \u3068 Simulink \u306f\u3001AI \u30e2\u30c7\u30eb\u306e\u30e9\u30a4\u30d5\u30b5\u30a4\u30af\u30eb\u5168\u4f53\u3092\u898f\u5236\u3059\u308b\u305f\u3081\u306e <a href=\"https:\/\/www.mathworks.com\/discovery\/mlops.html\">MLOps<\/a> \u30d7\u30ed\u30bb\u30b9\u3092\u81ea\u52d5\u5316\u3057\u3001\u672c\u756a\u74b0\u5883\u3078\u306e\u30c7\u30d7\u30ed\u30a4\u5f8c\u306b AI \u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc\u306e\u904b\u7528\u3092\u7dad\u6301\u304a\u3088\u3073\u76e3\u8996\u3059\u308b\u3053\u3068\u3082\u3067\u304d\u307e\u3059\u3002<\/p>\n<h6><\/h6>\n<p style=\"font-size: 18px;\"><b>\u307e\u3068\u3081<\/b><\/p>\n<p>MATLAB \u3068 Simulink \u306f\u3001\u30a8\u30f3\u30b8\u30cb\u30a2\u306b\u30c7\u30fc\u30bf\u54c1\u8cea\u306e\u5411\u4e0a\u3001\u4eba\u6c17\u306e\u3042\u308b\u6a5f\u68b0\u5b66\u7fd2\u304a\u3088\u3073\u30c7\u30a3\u30fc\u30d7\u30e9\u30fc\u30cb\u30f3\u30b0\u30a2\u30eb\u30b4\u30ea\u30ba\u30e0\u3078\u306e\u30a2\u30af\u30bb\u30b9\u3001\u30b7\u30b9\u30c6\u30e0\u5168\u4f53\u306e\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u74b0\u5883\u3068\u306e\u7d71\u5408\u3001\u305d\u3057\u3066\u5b9f\u4e16\u754c\u306e\u30b7\u30b9\u30c6\u30e0\u3078\u306e\u5c55\u958b\u306e\u305f\u3081\u306e\u30c4\u30fc\u30eb\u3092\u63d0\u4f9b\u3057\u307e\u3059\u3002\u305d\u306e\u305f\u3081\u3001\u6b63\u78ba\u306a\u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc\u3092\u8a2d\u8a08\u3059\u308b\u3060\u3051\u3067\u306a\u304f\u3001\u610f\u56f3\u3057\u305f\u30b7\u30b9\u30c6\u30e0\u3067\u52b9\u7387\u7684\u306b\u52d5\u4f5c\u3059\u308b\u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc\u3092\u8a2d\u8a08\u304a\u3088\u3073\u5c55\u958b\u3059\u308b\u305f\u3081\u306b\u3001\u6700\u9069\u306a AI \u6280\u8853\uff08\u307e\u305f\u306f\u4ed6\u306e\u65b9\u6cd5\u3068\u306e\u7d44\u307f\u5408\u308f\u305b\uff09\u3092\u9078\u629e\u3059\u308b\u3053\u3068\u304c\u3067\u304d\u307e\u3059\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"<div class=\"overview-image\"><img decoding=\"async\"  class=\"img-responsive\" src=\"https:\/\/blogs.mathworks.com\/deep-learning\/files\/2023\/08\/virtual_sensor_ai_workflow.png\" onError=\"this.style.display ='none';\" \/><\/div>\n<p>\u203b\u3053\u306e\u6295\u7a3f\u306f 2023 \u5e74 8 \u6708 11 \u65e5\u306b Artificial Intelligence \u3078 \u6295\u7a3f\u3055\u308c\u305f\u3082\u306e\u306e\u6284\u8a33\u3067\u3059\u3002<\/p>\n<p>&nbsp;<br \/>\n\u30d0\u30fc\u30c1\u30e3\u30eb\u30bb\u30f3\u30b5\u30fc\u306f\u3001\u30ea\u30a2\u30eb\u30bf\u30a4\u30e0\u306e\u76e3\u8996\u3068\u5236\u5fa1\u304c\u5fc5\u8981\u306a\u30b7\u30b9\u30c6\u30e0\u3067\u3001\u7269\u7406\u7684\u306a\u30bb\u30f3\u30b5\u30fc\u306e\u4f7f\u7528\u304c\u5b9f\u7528\u7684\u3067\u306a\u3044\u3001\u307e\u305f\u306f\u9ad8\u4fa1\u3067\u3042\u308b\u5834\u5408\u306b\u4f7f\u7528\u3067\u304d\u307e\u3059\u3002\u305d\u3057\u3066 AI&#8230; <a class=\"read-more\" href=\"https:\/\/blogs.mathworks.com\/japan-community\/2024\/07\/23\/virtual-sensors-with-ai-and-model-based-design-jp\/\">read more >><\/a><\/p>\n","protected":false},"author":159,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":[],"categories":[61],"tags":[],"_links":{"self":[{"href":"https:\/\/blogs.mathworks.com\/japan-community\/wp-json\/wp\/v2\/posts\/12027"}],"collection":[{"href":"https:\/\/blogs.mathworks.com\/japan-community\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/blogs.mathworks.com\/japan-community\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/blogs.mathworks.com\/japan-community\/wp-json\/wp\/v2\/users\/159"}],"replies":[{"embeddable":true,"href":"https:\/\/blogs.mathworks.com\/japan-community\/wp-json\/wp\/v2\/comments?post=12027"}],"version-history":[{"count":8,"href":"https:\/\/blogs.mathworks.com\/japan-community\/wp-json\/wp\/v2\/posts\/12027\/revisions"}],"predecessor-version":[{"id":12049,"href":"https:\/\/blogs.mathworks.com\/japan-community\/wp-json\/wp\/v2\/posts\/12027\/revisions\/12049"}],"wp:attachment":[{"href":"https:\/\/blogs.mathworks.com\/japan-community\/wp-json\/wp\/v2\/media?parent=12027"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/blogs.mathworks.com\/japan-community\/wp-json\/wp\/v2\/categories?post=12027"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/blogs.mathworks.com\/japan-community\/wp-json\/wp\/v2\/tags?post=12027"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}