An animated demonstration of an ODE agent session: a prompt asking for the safe fast-charge thermal envelope of a 96s2p battery pack sits in the composer, the agent replies with its approach, and four tool calls run in turn — build_model draws the pack's thermal network on the Modelica canvas, sweep fills a cluster card with returning cases, validate plots the temperature trace on a chart card, and create_document closes the session with a compiled paper and its files. A second run, optimising a 6-DOF robot arm trajectory, can be picked from the sidebar's Recent list.
物理世界をつくるエンジニアリングチームに選ばれています
ODE+ プラットフォーム
システムをモデル化し、実行できる Paper にして、その設計空間を探索する。すべてつながって。
Build & Simulate Agent
Component
物理システムをブラウザでモデリング＆シミュレーション。ポート検証つきのダイアグラムキャンバス、専用のランボードを備えたブラウザの中で動く本物のソルバー、そしてモデルを一緒に編集する AI コパイロット。
A Modelica diagram canvas, close in. A connection is being drawn from the motor's rotational flange flange_b toward the gearbox's flange_a, which has lit up as a compatible target; the gearbox's thermal heatPort and the motor's two electrical pins are dimmed at the same moment, because a rotational flange cannot connect to either.
Compute & Notebook Agent
Paper
実行できる計算論文。文章・モデル・プロットからなるノートブックで、ランタイムドックのライブ計算が走り、再現可能でファイルに依存しない記録が残ります。
Three blocks of one computational paper, side by side. A notebook paragraph concluding that at 4000 rpm the coolant outlet temperature stays below its 85 degree Celsius limit; the Python cell that computes it from cooler.port_b.T; and the chart that cell is still drawing, the temperature curve climbing and flattening below a dashed limit line.
Optimize & Explore Agent
Exploration
設計空間を大規模に探索。スイープ、最適化、感度分析＆サンプリング。平行座標で収束をライブに見ながら、計算クラスタからパレート解を受け取ります。
A parallel-coordinates plot of a gearbox ratio sweep across four axes: ratio, mass, efficiency and peak temperature. Of 250 candidate designs 168 have returned and sit faint in the background, while the fifteen Pareto-optimal ones lift out of them low on mass and high on efficiency; case 169 is still solving, so its line stops at the efficiency axis.
Model & CAD Agent
Design
ブラウザのための AI ネイティブ CAD ツール。AI ネイティブなオーサリングを Manifold と OCCT のカーネルの上に構築し、Modelica モデルと直結させます。
The drivetrain's road wheel on a CAD canvas, caught mid-rebuild: n_lugs has just gone from five to six, five holes are cut on the 114.3 mm bolt circle and the sixth is placed but not yet machined, beside the five parameters that drive the part.
コア機能
01
AI ネイティブ
02
プロ仕様の GUI
03
チームコラボレーション
04
クラスタコンピューティング
05
クラウドストレージ
06
Modelica & Python
07
コミュニティ主導
08
ゼロインストール、ブラウザネイティブ
つくる人たちに選ばれて
1 / 5
ODE+ is a fast-pace evolving platform that will enable “Modelica-driven engineering”. I am thrilled to be part of this fellowship program and contribute to expanding the reach of Modelica.
Dr. Clément Coïc
Technology Cluster Lead, Siemens Healthineers
I’m proud to represent Orthogonal as a fellow—being part of their journey of creativity, expertise, and innovation makes every collaboration exciting!
Daniel Ossmann
Full Professor for Control Systems, Munich University of Applied Sciences
Orthogonal provides a unique ecosystem that allows to leverage latent powers of Modelica, all by exploiting modern software innovations. I am excited of the opportunities for innovation that Orthogonal’s tools can unleash!
Luigi Vanfretti
Full Professor, Rensselaer Polytechnic Institute
ODE offers modern features such as cloud-native architecture for Modelica, FMI and SSP models with nicely integrated analysis features. It is very useful for my TUM lecture because students can access it directly without installing anything, and they can use it on all their devices, including tablets.
Hon. Prof. Dr.-Ing. Martin Otter
Former Modelica Association Chair
Opening up engineering platforms with browser-based technology is drastically lowering the entrance barrier especially for students and young engineers. With its rich support for equation-based modeling and FMI, ODE helps to proliferate the standards of the Modelica Association.
Dr-Ing. Dirk Zimmer
Chair of the Modelica Association
