<p>With the rapid development of vehicle technology and the advancement of smart city construction, vehicular networking has become an important part of future urban transportation systems. However, data transmission reliability and security challenges are faced in vehicular networking due to the high-speed movement of vehicles, dynamic changes in network topology and interference from malicious nodes. Malicious nodes may disrupt the network by tampering with data, forging information, or intentionally delaying transmission. In this paper, we propose a multi-path routing trust assessment model (PMT) combined with the Proof-of-Stake (PoS) mechanism, which dynamically calculates the trust value of nodes through the Pos mechanism, and designs the redundant path transmission mechanism to ensure the reliability of data packet transmission. Specifically, the vehicle trust value is based on the consensus algorithm of the amount of tokens pledged by the node, and the more the amount of tokens pledged, the higher the trust value of the node in the network. Our proposed redundant path transmission mechanism limits the number of low-trust paths when transmitting data and prioritizes paths with high trust values, which reduces the risk of redundant packets being attacked. We propose a malicious node detection mechanism that detects vehicles with trust values below a threshold as malicious nodes and consider them as attackers based on a trust assessment model. We have fully experimented in omnet++ simulation platform to simulate various attack scenarios, and the experimental results show that we can maintain more than 90% data delivery rate and achieve more than 95% Precision in malicious node detection.</p>