Implementation:Google deepmind Mujoco mj step
| Knowledge Sources | |
|---|---|
| Domains | Physics_Simulation, Numerical_Integration |
| Last Updated | 2026-02-15 06:00 GMT |
Overview
Concrete tool for advancing the MuJoCo simulation by one timestep provided by the MuJoCo C API.
Description
The mj_step function performs one complete simulation step: state validation, forward dynamics (via mj_forward), and time integration. The integrator is selected from the model options (Euler, RK4, or implicit). After calling this function, d->time is advanced by m->opt.timestep and all state variables are updated.
Usage
Call this function in a loop to run a simulation. Set d->ctrl before each call to apply control signals. The timestep is determined by m->opt.timestep (default 0.002 seconds).
Code Reference
Source Location
- Repository: mujoco
- File: src/engine/engine_forward.c
- Lines: 1422-1456
Signature
void mj_step(const mjModel* m, mjData* d);
Import
#include <mujoco/mujoco.h>
I/O Contract
Inputs
| Name | Type | Required | Description |
|---|---|---|---|
| m | const mjModel* | Yes | Compiled physics model (immutable) |
| d | mjData* | Yes | Simulation state with control signals in d->ctrl |
Outputs
| Name | Type | Description |
|---|---|---|
| d (modified) | mjData* | Advanced state: updated qpos, qvel, time, contacts, forces |
Usage Examples
Basic Simulation Loop
#include <mujoco/mujoco.h>
// Load and initialize
mjModel* m = mj_loadXML("humanoid.xml", NULL, error, 1000);
mjData* d = mj_makeData(m);
// Simulate for 10 seconds
while (d->time < 10.0) {
// Set control signals
d->ctrl[0] = 1.0; // actuator 0
// Advance simulation by one timestep
mj_step(m, d);
}
printf("Final time: %f\n", d->time);
mj_deleteData(d);
mj_deleteModel(m);
Related Pages
Implements Principle
Requires Environment
Uses Heuristic
Page Connections
Double-click a node to navigate. Hold to expand connections.
Principle
Implementation
Heuristic
Environment