Roland and Massing, 1971. 
I do not think his equations are too difficult. His derivations is quite 
straight-forward. Up to and including Section 2.5, the analysis is basically 
sound. There are some slips of the pen, however. Most of the slips only 
affect the non-linear terms. In his figures, the actual signs of some 
dimensions are negative in the configuration shown; this can be confusing. 
The slips in the equations of motion that I found are: 

page 11: in \gamma_{22}, 
second line, first term: there should be a factor 2 before it. 
third line: the cosine-squared should not be there.  

page 13, equation of motion (2.3.21): 
In the right-hand side, the first term should have a minus sign and the second 
a plus sign. In the very last term, the - x_F'' should not be there (this 
affects the linear terms). 

page 16, (2.4.12), expression for \gamma_{13}: 
first line, sign last term but one, -y_D pq 
second line, first term: sign again; second term: the 2 should not be there. 

page 16, (2.4.13), expression for \gamma_{23}: 
first line: sign: +y_D(p^2 + q^2)  

page 18, equation of motion (2.4.21): 
first two terms in the right-hand side: signs should be - in both of them. 

page 22, (2.5.20), equation of motion for rotation about the X axis: 
fourth line, the last term before \dot{q} should not be there, but instead 
before \dot{r} in the following expression between ( ) with plus sign, because 
the minus is before the ( ). This affects the linear equations. 
seventh line: the \ddot{\phi}_D is missing before the equal sign. This is 
corrected in the matrix expression on page 37. 
eighth line, fourth term: the sign should be +, so +\cos ... 
This affects the linear equations. 

page 22, (2.5.21), equation of motion for rotation about the Y axis: 
first line, second term: \dot{v} should be \dot{w} 
second line, second term: \rho_D should be y_D . 
third line: y_F should be z_F (two times) 
I think that the expression [1 - \cos \phi_D] (three places) should be 
simply zero. This term is the consequence of some simplifications for small 
roll angles of the rider made earlier. 
eighth line: there should be squares at the cos and sin 
ninth line: second term should have a plus sign. 

page 23, (2.5.22), equation of motion for rotation about the Z axis: 
third line, last but one term: y_F should be squared. 
sixth line, first term after plus: a term \cos\delta has been forgotten. 
It is there on Page 37. 
ninth line: squares again at cos and sin of \phi_D. 
tenth line: an I_{DXY} has been forgotten. 

There are further slips is the intermediate results. 

These were just slips, and not essential errors. In Section 2.6, however, 
an approximation in the calculation of the side slip angles is made. For the 
rear wheel, this seems to be correct for the linearized results. For the 
front wheel, however, no term due to \dot{\delta} is included. So if we 
put Roland's slip angles equal to zero and consider only linearized 
equations, our results do not agree. However, if we use the correct linear 
expression for the slip angle, make the corrections listed above, 
linearize the equations and eliminate the constraint forces, our  
results agree. (after some 20 pages of calculations)  

Roland (1973) gives basically the same equations. Apparently, no typos were 
corrected, only some further typos were introduced. 

JPM, Nottingham, 8 Nov. 2005.

Ik ben nu in Roland and Massing (1971) aan het lezen. Er zitten daar nogal wat 
'slips of the pen' in. Ook de sliphoek van het voorwiel is niet exact voor het 
lineaire geval. Zijn latere artikel uit 1973 (Roland 1973) heb ik niet. 
Daar schijnt een en ander verbeterd te zijn. Er zijn dus niet alleen 
drukfouten, maar ook echte fouten/benaderingen. Kan je me een afdrukje van 
(Roland 1973) sturen? Overigens is hun meetfiets wel interessant. Je bent 
echt niet de eerste met je fiets. 

JPM, Nottingham, 2 Nov. 2005.