With　the　rapid　deployment　of　the　Internet　of　Vehicles　(IoV),　user’s　requirements　for　service　quality　of　network　are　also　increasing.　As　one　of　the　important　contents　of　IoV　network　service,　data　computing　has　attracted　more　attention.　Mobile　edge　computing　(MEC)　allows　vehicles　to　offload　computing　tasks　to　MEC　servers　at　the　edge　of　IoV　system,　so　as　to　decrease　computing　delay　and　improve　efficiency　effectively.　However,　there　are　some　essential　issues　to　be　considered:　(1)　a　rapid　increase　in　the　demand　for　edge　computing　devices,　and　(2)　privacy　and　security　in　the　process　of　data　transmission　and　sharing.　This　paper　proposes　a　blockchain-based　parking-vehicle-assisted　computing　system　model　for　the　scenario　of　mobile　vehicle　offloading　in　IoV.　Based　on　joint　consideration　of　computing　resources　and　vehicle　mobility,　deep　reinforcement　learning　(DRL)　is　used　to　reduce　system　energy　consumption,　data　transmission　delay,　and　improve　blockchain　throughput　by　optimizing　computing　offloading　and　resource　allocation　strategies.　The　simulation　results　show　that　compared　with　other　existing　schemes　the　proposed　optimization　scheme　can　improve　the　system　performances　effectively　and　significantly.　©　2023　Inst.　of　Scientific　and　Technical　Information　of　China.　All　rights　reserved.