Public research sourceUC San DiegoHao Su Lab / SAPIEN
UCSD ManiSkill: Teaching Many Simulated Robots at Once
First understand what UC San Diego researchers are trying to learn through Hao Su Lab / SAPIEN, why the question matters, how they test it, and what the result cannot prove. Let many simulated robots practise picking up a cube at once and compare how training conditions change the result. The student then completes A robot-learning experiment dashboard that batch-tests training conditions and compares failure scenes. For a college application or interview, the student separates the source research from their feature, interface, tests, failures, and revisions. Available online or in person in Apgujeong, in one-to-one or small-group formats.
University and lab names show where each public project came from. CIT independently designed these courses; they are not official university courses, partnerships, or endorsements.
8guided sessions
4-10 hoursto the first small project
Beginner-intermediate, advanced for trainingrecommended level
Not requiredphysical robot
Primary routes
Choose the student's primary AI education and portfolio route
CIT keeps the international-school route first, followed by the separate science and gifted-school route.
CIT lessons can run online across Korea and overseas or in person in Apgujeong. The student owns the question, code, tests and explanation; classes do not guarantee admission, selection or awards, and online delivery never changes an institution's submission rules.
Does training many copies of the same simulated robot at once help it learn faster?
Train many simulated robots at the same time and compare how object size, friction, and goal position affect what they learn.
Students first use a small working example to understand the researchers' question, method, and evidence. A controlled change helps identify what the student's extension must solve. The student then implements a useful feature, connects an operator interface, and tests the integrated application in normal, boundary, and failure cases.
What will the student complete?
A robot-learning experiment dashboard that batch-tests training conditions and compares failure scenes
The final package includes runnable instructions, the feature and interface design, normal and failure tests, one documented revision, and a three-minute explanation in the student's own words.
Feature
A feature that creates and tests batches of object-size, friction, and goal-position combinations to find the weakest condition
Operator interface
An experiment dashboard comparing training settings, practice count, success rate, and failed scenes side by side
Optional LLM boundary
Optional: an LLM proposes a next experiment using only selected graphs and run records, without changing training settings automatically.
Integrated result
A robot-learning experiment dashboard that batch-tests training conditions and compares failure scenes
Four ideas explained in this course
01many robots practising at once
02random changes to practice conditions
03numbers versus camera images
04scores that reward success
Eight introductory sessions
Understand the research, then build and test a working robotics application
The 20 courses are eight-session CIT studios in which students understand a public university or lab project, confirm a working example, and then turn it into a small robotics application with a useful feature and an operator interface. Selected courses add an optional LLM explanation tool that can read run records but cannot control the robot. RoboMaster has 28 sessions, and implementation scope is adjusted to each student's experience and computer access.
Understand the source
01
Run the working cube-pickup example
02
Understand the simulated robot, object, and actions
Measure and compare
03
Compare numeric information with camera images
04
Compare a fixed rule with random actions
Build a feature
05
Change object size, friction, and goal position
06
Implement the student-owned feature: A feature that creates and tests batches of object-size, friction, and goal-position combinations to find the weakest condition
Integrate and demonstrate
07
Build the operator interface: An experiment dashboard comparing training settings, practice count, success rate, and failed scenes side by side. Add the bounded assistant: Optional: an LLM proposes a next experiment using only selected graphs and run records, without changing training settings automatically.
08
Integrate, test, and demonstrate: A robot-learning experiment dashboard that batch-tests training conditions and compares failure scenes
Public project used in class
See the exact version CIT reviewed
Public projects can change over time. To keep the class example consistent, CIT uses version 62ff3a5896b4 of mani-skill/ManiSkill. CIT checked it on 2026-08-14; it was created on 2026-08-02. Usage-rights note: Apache-2.0.
The student changes only the number of practice scenes and checks whether faster also means better. In class the project is separated into five boxes (input, memory, process, output, control). The lesson opens one of them, changes one value, and leaves the rest closed.
InputWhat comes in?
The task definition that says what counts as success, and what the cameras send back
Memoryopened hereWhat persists?
The state where hundreds of scenes' positions and speeds sit together in one array
ProcessWhat transforms?
Turning a command into motion, then scoring how close the result came to the task
OutputWhat leaves, and who uses it?
The observation, the score and the success flag that come back for each scene
ControlWhat decides when anything runs?
What decides how many scenes run together, and when a failed scene is reset
The files the lesson opens, by name. The course is not a walk through the repository; it opens a chosen few and says which.
One value moves and everything else stays. The right-hand column is the prediction written before the run, not a result; the work is reconciling the two.
The universities and labs named here made the open projects this course reads. CIT designed the course independently; it is not an official, affiliated, or endorsed course.
What the student needs
Computer
CPU simulation possible, GPU recommended for parallel training
Clear answers about what students do, what they need, and where the project came from.
Is UCSD ManiSkill: Teaching Many Simulated Robots at Once an official course from UC San Diego Hao Su Lab and SAPIEN?
University and lab names show where each public project came from. CIT independently designed these courses; they are not official university courses, partnerships, or endorsements.
Is hardware required?
Not required. The course starts with a robot on the computer or a saved recording of a completed run. Computer guidance: CPU simulation possible, GPU recommended for parallel training.
What background should a student have?
Recommended level: Beginner-intermediate, advanced for training. Students should be able to follow a guided Python example, test one controlled change, and then build and explain a small feature and interface.
What will the student make?
The completed project is A robot-learning experiment dashboard that batch-tests training conditions and compares failure scenes. The student implements A feature that creates and tests batches of object-size, friction, and goal-position combinations to find the weakest condition and An experiment dashboard comparing training settings, practice count, success rate, and failed scenes side by side. Optional: an LLM proposes a next experiment using only selected graphs and run records, without changing training settings automatically.
How can the student use this project in a college application?
Explain the source research question, method, evidence, and limits first. Then separate the student's own feature and interface decisions, normal and failure tests, revisions, and next question. The source institution's name does not imply affiliation or guarantee admission.
Which project version does the course use?
CIT reviewed version 62ff3a5896b4 of mani-skill/ManiSkill on 2026-08-14. That version was created on 2026-08-02. We keep this version during class so the example does not change unexpectedly, and we check the setup again before teaching.
Can UCSD ManiSkill: Teaching Many Simulated Robots at Once be taken online or in person, one-to-one or in a small group?
Yes. This course is offered online and in person at CIT in Apgujeong, Gangnam-gu, Seoul, with one-to-one and small-group options. Placement and current availability are confirmed after a readiness consultation.
Is this a good first project for this student?
Before placement, we check the student's coding and math experience, available computer, interests, and ability to explain what happened.
Stanford robomimic: Learning Robot Actions from DemonstrationsReplay robot actions demonstrated by people, then compare how the number of examples and camera information...→
Georgia Tech MimicLabs: Creating Robot Practice DataTurn a small number of robot demonstrations into many practice scenes, then check whether the larger set adds...→
Princeton AdaptSim: Closing the Simulation-to-Reality GapA simulated robot behaves differently when its weight or friction does not match reality. Test which adjustment...→