IoT　routing　is　the　basis　for　efficient　operation　of　perception　layer　networks.　The　current　routing　technology　of　IoT　lacks　a　trusted　authentication　mechanism,　and　the　adopted　traditional　routing　algorithm　can　hardly　meet　the　routing　demand　of　massive　IoT　terminals.　Based　on　this,　this　paper　proposes　a　trusted　authentication　mechanism　for　IoT　routing,　and　combines　reinforcement　learning　and　trickle　timer　to　propose　an　intelligent　adaptive　trickle　timer　algorithm,　which　ensures　the　secure　and　trustworthy　and　efficient　operation　of　IoT　routing.　©　2022　ACM.