Algorithm 1: Abnormal road surface recognition method

Input: z, the Z-axis acceleration; v, the vehicle speed; KS, the spring stiffness; CS, the damping coefficient.   Output: event_z, Acceleration due to abnormal road surface.   1. Algorithm begin:   2. μ = 0   % μ is a mathematical expectation   3. σ = 0   % σ is a standard deviation   4. TG = 2  % TG is the Gaussian matching threshold   5. TV = 20  % TV is the speed threshold   6. if (v > TV)   7.  z_match ←abs(z − μ)/σ   8. % calculating thresholds TZ using fuzzy logic inference machines   9.  TZ← fuzzy_control (KS, CS)   10.  if (z_match > TG*v/TV) && (abs(z) > (TZ* v/TV))   11.   event_z ← z   12.  else    % update μ and σ, as described in Equation (5)   13.   μ ← (1 − α) * μ + α*z   14.   σ ← SQRT ((1 − α) * σ^2 + α * (z − μ)^2)  % SQRT means calculate square root   15.  end if   16. end if   17. return event_z   18. Algorithm end