diff --git a/sw/airborne/math/pprz_simple_matrix.h b/sw/airborne/math/pprz_simple_matrix.h
index 4b6fd9be8b..b74aa6dd26 100644
--- a/sw/airborne/math/pprz_simple_matrix.h
+++ b/sw/airborne/math/pprz_simple_matrix.h
@@ -13,38 +13,38 @@
 //
 // C = A*B   A:(i,k) B:(k,j) C:(i,j)
 //
-#define MAT_MUL(_i, _k, _j, C, A, B) {				\
-    int l,c,m;							\
-    for (l=0; l<_i; l++)					\
-      for (c=0; c<_j; c++) {					\
-	C[l][c] = 0.;						\
-	for (m=0; m<_k; m++)					\
-	  C[l][c] += A[l][m]*B[m][c];				\
-      }								\
+#define MAT_MUL(_i, _k, _j, C, A, B) {          \
+    int l,c,m;                                  \
+    for (l=0; l<_i; l++)                        \
+      for (c=0; c<_j; c++) {                    \
+        C[l][c] = 0.;                           \
+        for (m=0; m<_k; m++)                    \
+          C[l][c] += A[l][m]*B[m][c];           \
+      }                                         \
   }
 
 //
 // C = A*B'   A:(i,k) B:(j,k) C:(i,j)
 //
-#define MAT_MUL_T(_i, _k, _j, C, A, B) {			\
-    int l,c,m;							\
-    for (l=0; l<_i; l++)					\
-      for (c=0; c<_j; c++) {					\
-	C[l][c] = 0.;						\
-	for (m=0; m<_k; m++)					\
-	  C[l][c] += A[l][m]*B[c][m];				\
-      }								\
+#define MAT_MUL_T(_i, _k, _j, C, A, B) {        \
+    int l,c,m;                                  \
+    for (l=0; l<_i; l++)                        \
+      for (c=0; c<_j; c++) {                    \
+        C[l][c] = 0.;                           \
+        for (m=0; m<_k; m++)                    \
+          C[l][c] += A[l][m]*B[c][m];           \
+      }                                         \
   }
 
 
 //
 // C = A-B
 //
-#define MAT_SUB(_i, _j, C, A, B) {					\
-    int l,c;								\
-    for (l=0; l<_i; l++)						\
-      for (c=0; c<_j; c++)						\
-	C[l][c] = A[l][c] - B[l][c];					\
+#define MAT_SUB(_i, _j, C, A, B) {              \
+    int l,c;                                    \
+    for (l=0; l<_i; l++)                        \
+      for (c=0; c<_j; c++)                      \
+        C[l][c] = A[l][c] - B[l][c];            \
   }
 
 
@@ -53,31 +53,31 @@
 //
 // invS = 1/det(S) com(S)'
 //
-#define MAT_INV33(_invS, _S) {						\
-    const float m00 = _S[1][1]*_S[2][2] - _S[1][2]*_S[2][1];		\
-    const float m10 = _S[0][1]*_S[2][2] - _S[0][2]*_S[2][1];		\
-    const float m20 = _S[0][1]*_S[1][2] - _S[0][2]*_S[1][1];		\
-    const float m01 = _S[1][0]*_S[2][2] - _S[1][2]*_S[2][0];		\
-    const float m11 = _S[0][0]*_S[2][2] - _S[0][2]*_S[2][0];		\
-    const float m21 = _S[0][0]*_S[1][2] - _S[0][2]*_S[1][0];		\
-    const float m02 = _S[1][0]*_S[2][1] - _S[1][1]*_S[2][0];		\
-    const float m12 = _S[0][0]*_S[2][1] - _S[0][1]*_S[2][0];		\
-    const float m22 = _S[0][0]*_S[1][1] - _S[0][1]*_S[1][0];		\
-    float det = _S[0][0]*m00 - _S[1][0]*m10 + _S[2][0]*m20;   	\
-    if (fabs(det) < FLT_EPSILON) {				\
+#define MAT_INV33(_invS, _S) {                                          \
+    const float m00 = _S[1][1]*_S[2][2] - _S[1][2]*_S[2][1];            \
+    const float m10 = _S[0][1]*_S[2][2] - _S[0][2]*_S[2][1];            \
+    const float m20 = _S[0][1]*_S[1][2] - _S[0][2]*_S[1][1];            \
+    const float m01 = _S[1][0]*_S[2][2] - _S[1][2]*_S[2][0];            \
+    const float m11 = _S[0][0]*_S[2][2] - _S[0][2]*_S[2][0];            \
+    const float m21 = _S[0][0]*_S[1][2] - _S[0][2]*_S[1][0];            \
+    const float m02 = _S[1][0]*_S[2][1] - _S[1][1]*_S[2][0];            \
+    const float m12 = _S[0][0]*_S[2][1] - _S[0][1]*_S[2][0];            \
+    const float m22 = _S[0][0]*_S[1][1] - _S[0][1]*_S[1][0];            \
+    float det = _S[0][0]*m00 - _S[1][0]*m10 + _S[2][0]*m20;             \
+    if (fabs(det) < FLT_EPSILON) {                                      \
       /* If the determinant is too small then set it to epsilon preserving sign. */ \
       warn_message("warning: %s:%d MAT_INV33 trying to invert non-invertable matrix '%s' and put result in '%s'.\n", __FILE__, __LINE__, #_S, #_invS); \
-      det = copysignf(FLT_EPSILON, det);  \
-    }                                     \
-    _invS[0][0] =  m00 / det;						\
-    _invS[1][0] = -m01 / det;						\
-    _invS[2][0] =  m02 / det;						\
-    _invS[0][1] = -m10 / det;						\
-    _invS[1][1] =  m11 / det;						\
-    _invS[2][1] = -m12 / det;						\
-    _invS[0][2] =  m20 / det;						\
-    _invS[1][2] = -m21 / det;						\
-    _invS[2][2] =  m22 / det;						\
+      det = copysignf(FLT_EPSILON, det);                                \
+    }                                                                   \
+    _invS[0][0] =  m00 / det;                                           \
+    _invS[1][0] = -m01 / det;                                           \
+    _invS[2][0] =  m02 / det;                                           \
+    _invS[0][1] = -m10 / det;                                           \
+    _invS[1][1] =  m11 / det;                                           \
+    _invS[2][1] = -m12 / det;                                           \
+    _invS[0][2] =  m20 / det;                                           \
+    _invS[1][2] = -m21 / det;                                           \
+    _invS[2][2] =  m22 / det;                                           \
   }