Jump to content

Connect SuperML | Leeroopedia MCP: Equip your AI agents with best practices, code verification, and debugging knowledge. Powered by Leeroo — building Organizational Superintelligence. Contact us at founders@leeroo.com.

Implementation:Google deepmind Mujoco MjTnum Header

From Leeroopedia
Knowledge Sources
Domains C API, Numeric Types, Precision Configuration, Type Definitions
Last Updated 2026-02-15 04:00 GMT

Overview

Compact C header file that defines MuJoCo's fundamental numeric type (mjtNum) with compile-time single/double precision selection, byte type, and size type.

Description

mjtnum.h is a minimal 46-line header that establishes the core numeric types used throughout MuJoCo. By default, mjtNum is typedef'd to double with mjMINVAL set to 1E-15 for denominator safety. When mjUSESINGLE is defined at compile time, mjtNum becomes float with mjMINVAL as 1E-15f. The header also defines mjtByte (unsigned char, used for boolean flags) and mjtSize (int64_t, used for buffer sizes). These three types form the numeric foundation that all other MuJoCo headers and computation code depend on.

Usage

This header is included by mjmodel.h and transitively by all MuJoCo code. It is used whenever MuJoCo's floating-point precision needs to be configurable between single and double precision at compile time. The mjtByte type is used for boolean arrays in model data, and mjtSize is used for memory allocation sizes.

Code Reference

Source Location

Key Functions

// Floating-point type with compile-time precision selection
#ifndef mjUSESINGLE
  typedef double mjtNum;
  #define mjMINVAL    1E-15       // minimum value in any denominator
#else
  typedef float mjtNum;
  #define mjMINVAL    1E-15f
#endif

// Byte type for boolean flags
typedef unsigned char mjtByte;    // used for true/false

// Size type for buffer allocation
typedef int64_t mjtSize;          // used for buffer sizes

Import

#include <mujoco/mjtnum.h>

I/O Contract

Inputs

Name Type Required Description
mjUSESINGLE preprocessor define No If defined, switches mjtNum from double to float precision

Outputs

Name Type Description
mjtNum typedef (double or float) Primary floating-point type used for all MuJoCo numerical computations
mjMINVAL #define (1E-15 or 1E-15f) Minimum safe denominator value to prevent division by zero
mjtByte typedef (unsigned char) Byte-sized type used for boolean flag arrays
mjtSize typedef (int64_t) 64-bit integer type used for buffer and memory sizes

Related Pages

Page Connections

Double-click a node to navigate. Hold to expand connections.
Principle
Implementation
Heuristic
Environment