Some question about configurations

[TK72]

I run the VOXELMAP program with kitti No.0, No.6, No.7 dataset. But I didn't get the same ATE results in the paper. I'm wondering if my configuration file is not correct.

My velodyne.yaml is written as follows:
common:
lid_topic: "/kitti/velo/pointcloud"
imu_topic: "/no_imu"

preprocess:
lidar_type: 2 #1 for Livox serials LiDAR, 2 for Velodyne LiDAR, 3 for L515 LiDAR
scan_line: 64
blind: 1
point_filter_num: 1
calib_laser: true # true for KITTI Odometry dataset

mapping:
down_sample_size: 0.5
max_iteration: 3
voxel_size: 3.0
max_layer: 4 # 4 layer, 0, 1, 2, 3
layer_point_size: [5, 5, 5, 5, 5]
plannar_threshold: 0.01
max_points_size: 1000
max_cov_points_size: 1000

noise_model:
ranging_cov: 0.04
angle_cov: 0.1
acc_cov_scale: 1.0
gyr_cov_scale: 0.5

imu:
imu_en: false
extrinsic_T: [ 0, 0, 0]
extrinsic_R: [ 1, 0, 0,
0, 1, 0,
0, 0, 1]

visualization:
pub_voxel_map: false
publish_max_voxel_layer: 1 # only publish 0,1,2 layer's plane
pub_point_cloud: false
dense_map_enable: false
pub_point_cloud_skip: 5 # publish one points per five points

Result:
write_kitti_log: true
result_path: "/home/trajectory.txt"

//////////////////////////////////////////////////////////
I extract the lidar pose with
inline void kitti_log(FILE *fp) {
Eigen::Matrix4d T_lidar_to_cam;
T_lidar_to_cam << 0.00042768, -0.999967, -0.0080845, -0.01198, -0.00721062,
0.0080811998, -0.99994131, -0.0540398, 0.999973864, 0.00048594,
-0.0072069, -0.292196, 0, 0, 0, 1.0;
V3D rot_ang(Log(state.rot_end));
MD(4, 4) T;
T.block<3, 3>(0, 0) = state.rot_end;
T.block<3, 1>(0, 3) = state.pos_end;
T(3, 0) = 0;
T(3, 1) = 0;
T(3, 2) = 0;
T(3, 3) = 1;

//output camera pose not lidar pose
T = T_lidar_to_cam * T * T_lidar_to_cam.inverse();
//output TUM format result
Eigen::Quaterniond q(state.rot_end);
fprintf(fp, "%lf %lf %lf %lf %lf %lf %lf %lf\r\n", Measures.lidar_beg_time - first_lidar_time,
state.pos_end[0], state.pos_end[1], state.pos_end[2], q.x(), q.y(), q.z(), q.w());
/* fprintf(fp, "%lf %lf %lf %lf %lf %lf %lf %lf \r\n", Measures.lidar_beg_time - first_lidar_time,
state.pos_end[0], state.pos_end[1], state.pos_end[2], q.w(), q.x(), q.y(), q.z());*/
fflush(fp);
}

//////////////////////////////////////////
kitti_00_change.txt
///////////////

And I compared it with the real data kitti_00_change.txt using evo_ape function.
I got about 6 meters ATE in average.

I'm wondering if I have made some mistakes in the comparing steps.