基于可达集和强化学习的智能汽车决策规划

高洪伟,尚秉旭,张鑫康,王洪峰,何维,裴晓飞

Decision-making and planning of intelligent vehicle based on reachable set and reinforcement learning

Hongwei GAO,Bingxu SHANG,Xinkang ZHANG,Hongfeng WANG,Wei HE,Xiaofei PEI

表 1 横向需搜索的谓词

Tab.1 Required horizontal search predicates

决策出的驾驶行为 需搜索谓词 ${a_{{j}}}$ ${{\mathrm{if}}}{\mkern 1mu} {\mkern 1mu} {L_1}{\mkern 1mu} {\mkern 1mu} {\rm{Inlanelets}}({\rm{Ve}}{{\rm{h}}_0},\left\{ {{L_1}} \right\})$ ${\rm{Inlanelets}}({\rm{Ve}}{{\rm{h}}_0},\left\{ {{L_0}} \right\})$ ${{\mathrm{if}}}{\mkern 1mu} {\mkern 1mu} {L_{ - 1}}{\mkern 1mu} {\mkern 1mu} {\rm{Inlanelets}}({\rm{Ve}}{{\rm{h}}_0},\left\{ {{L_{ - 1}}} \right\})$ ${\mathrm{Lcf}}$ $\begin{array}{*{20}{c}} {{\rm{Inlanelets}}({\rm{Ve}}{{\rm{h}}_0},({L_1},{L_0}))} \\ {{\rm{Inlanelets}}({\rm{Ve}}{{\rm{h}}_0},({L_1}))} \\ {{\rm{Inlanelets}}({\rm{Ve}}{{\rm{h}}_0},({L_0}))} \end{array}$ ${\mathrm{Lcr}}$ $\begin{array}{*{20}{c}} {{\mathrm{{lnlanelets}}}({\rm{Ve}}{{\rm{h}}_0},\left\{ {{L_0},{L_1}} \right\})} \\ {{\mathrm{{Inlanelets}}}({\rm{Ve}}{{\rm{h}}_0},({L_{ - 1}}))} \\ {{\mathrm{{Inlanelets}}}({\rm{Ve}}{{\rm{h}}_0},({L_0}))} \end{array}$