Abstract: Due to the lack of consideration of long-range constraints in existing methods for evaluating the credibility of lidar data, resulting in inaccurate evaluation results, this paper proposes a design of a lidar sensor data credibility evaluation method. Firstly, an analysis of the lidar operating range is conducted on the lidar data. Then, using this as a constraint condition and combined with a BP neural network, a lidar sensing data evaluation model is established. Finally, the activation function of the hidden layer is used to activate the data′s credible significance, and the channel error and the error of the data to be evaluated are obtained through weight training. Based on the error threshold between the two, the data credibility was evaluated. The test results showed that the proposed method has high accuracy and strong feasibility in credibility evaluation, and the algorithm has certain adaptability.