Path Planning Algorithm for Unmanned Surface Vessel Based on Multiobjective Reinforcement Learning

<table class="table-group" id="tab3"><tr><td><table class="table"><tr><td class="thead-hr" colspan="5"><hr/></td></tr><tr class="thead"><td class="align_left">Method of policy integration</td><td class="align_center">Success rate of path planning (%)</td><td class="align_center">Success rate of goal approaching (%)</td><td class="align_center">Success rate of dynamic obstacle avoidance (%)</td><td class="align_center">Length of the path (m)</td></tr><tr><td class="thead-hr" colspan="5"><hr/></td></tr><tr><td class="align_left">Rank voting</td><td class="align_center">52</td><td class="align_center">55</td><td class="align_center">91</td><td class="align_center">17.68</td></tr><tr><td class="align_left">Linear voting</td><td class="align_center">57</td><td class="align_center">63</td><td class="align_center">94</td><td class="align_center">16.98</td></tr><tr><td class="align_left">Proposed method</td><td class="align_center">93</td><td class="align_center">96</td><td class="align_center">97</td><td class="align_center">17.96</td></tr><tr class="table-tr"><td colspan="5"><hr class="tbody-hr"/></td></tr></table></td></tr></table>

<div>Performance of various ensemble learning methods.</div>

Computational Intelligence and Neuroscience

tab3

Table 3

Table 3: Path Planning Algorithm for Unmanned Surface Vessel Based on Multiobjective Reinforcement Learning 