A Novel Particle Swarm Optimization Algorithm Based on Reinforcement Learning Mechanism for AUV Path Planning

<table class="algorithm-group"><tr><td><table class="algorithm" id="alg2"><tr><td>(1)</td><td>Initialize: the particles’ positions, velocity (V), the number of particles (P), maximum number of iterations (N), the global optimum (gbest), the individual optimum (pbest (i))</td></tr><tr><td>(2)</td><td>for maximum number of iterations (N)</td></tr><tr><td>(3)</td><td> for all particles (i)</td></tr><tr><td>(4)</td><td>  Update the position and V according to (<a href="https://static-preview.hindawi.com/articles/complexity/volume-2021/8993173/figures/#EEq14">14</a>)</td></tr><tr><td>(5)</td><td>  Calculate the fitness value of particles F (i) according to (<a href="https://static-preview.hindawi.com/articles/complexity/volume-2021/8993173/figures/#EEq11">11</a>)</td></tr><tr><td>(6)</td><td>  if (F (i) &lt; pbest (i)) then</td></tr><tr><td>(7)</td><td>   pbest (i) = F (i)</td></tr><tr><td>(8)</td><td>  end if</td></tr><tr><td>(9)</td><td>  if (pbest (i) &lt; gbest) then</td></tr><tr><td>(10)</td><td>   gbest = pbest (i)</td></tr><tr><td>(11)</td><td>  end if</td></tr><tr><td>(12)</td><td>  update ω (i) according to (<a href="https://static-preview.hindawi.com/articles/complexity/volume-2021/8993173/figures/#EEq13">13</a>)</td></tr><tr><td>(13)</td><td> end for</td></tr><tr><td>(14)</td><td>end for</td></tr></table></td></tr></table>

Complexity

alg2

Algorithm 2

Algorithm 2: A Novel Particle Swarm Optimization Algorithm Based on Reinforcement Learning Mechanism for AUV Path Planning 