ros运行SRR和SPP_RRR模式段错误

[936384885xy]

非常感谢你们能开源这个项目，我在jetson agx orin平台上进行了部署测试，能正常运行SPP配置，但是运行视觉融合的配置会报段错误。崩溃的位置是在 debug运行发现这里frame确实是空的，可以确定输入图像数据确实正常发布了，rviz可以正常显示图像。不知道我是哪里操作有误。 我的SRR配置文件内容如下：

GICI offline pseudo-real-time GNSS/IMU/Camera SRR estimation with ROS topics as input and output

stream: streamers:

• streamer: tag: str_ros_gnss_rov type: ros topic_name: /gici/gnss_rover io: input format: gnss_raw enable_observation: true
• streamer: tag: str_ros_gnss_ref type: ros topic_name: /gici/gnss_reference io: input format: gnss_raw enable_observation: true enable_antenna_position: true
• streamer: tag: str_ros_gnss_eph type: ros topic_name: /gici/gnss_ephemeris io: input format: gnss_raw enable_ephemeris: true enable_ionosphere_parameter: true
• streamer: tag: str_dcb_file output_tags: [fmt_dcb_file] type: file path: /home/l1/gici-open/option/CAS0MGXRAP_20221580000_01D_01D_DCB.BSX enable_time_tag: false
• streamer: tag: str_ros_imu type: ros topic_name: /gici/imu_raw io: input format: imu
• streamer: tag: str_ros_camera type: ros topic_name: /gici/image_raw io: input format: image
• streamer: tag: str_rtk_solution_file input_tags: [fmt_rtk_solution_file] type: file path: /home/l1/gici-open/output/rtk_solution.txt enable_time_tag: false
• streamer: tag: str_srr_solution_file input_tags: [fmt_srr_solution_file] type: file path: /home/l1/gici-open/output/srr_solution.txt enable_time_tag: false
• streamer: tag: str_solution type: ros topic_name: solution queue_size: 5 io: output format: pose_stamped
• streamer: tag: str_solution_odometry type: ros topic_name: solution_odometry queue_size: 10 io: output format: odometry subframe_id: Body
• streamer: tag: str_solution_path type: ros topic_name: solution_path queue_size: 100 io: output format: path
• streamer: tag: str_featured_image type: ros topic_name: featured_image queue_size: 3 io: output format: image

• streamer: tag: str_landmarks type: ros topic_name: landmarks queue_size: 3 marker_scale: 0.2 io: output format: marker

  formators:

• formator: io: input tag: fmt_dcb_file type: dcb-file
• formator: io: output tag: fmt_rtk_solution_file type: nmea use_esa: true use_esd: true talker_id: GP

• formator: io: output tag: fmt_srr_solution_file type: nmea use_esa: true talker_id: GP

  replay: enable: false

estimate:

• estimator: tag: est_rtk type: rtk input_tags: [str_ros_gnss_rov, str_ros_gnss_ref, str_ros_gnss_eph, fmt_dcb_file] str_ros_gnss_rov_roles: [rover] str_ros_gnss_ref_roles: [reference] str_ros_gnss_eph_roles: [ephemeris] fmt_dcb_file_roles: [code_bias] output_tags: [fmt_rtk_solution_file] output_align_tag: str_ros_gnss_rov compute_covariance: true rtk_options: max_window_length: 3 use_ambiguity_resolution: true estimate_velocity: true gnss_estimator_base_options: gnss_common: min_elevation: 12.0
  min_SNR: [35.0, 30.0] system_exclude: [R, E, C] code_exclude: [G|5Q] mw_slip_thres: 0.5 gf_slip_thres: 0.05 gf_sd_slip_thres: 0.05 use_outlier_rejection: true max_pesudorange_error: 4.0
  max_phaserange_error: 0.06
  max_doppler_error: 0.5 estimator_base_options: max_iteration: 5 num_threads: 2 max_solver_time: 0.02 verbose_output: false solver_type: dense_schur trust_region_strategy_type: dogleg ambiguity_resolution_options: system_exclude: [R] min_percentage_fixation_uwl: 1.0 min_percentage_fixation_wl: 0.9 min_percentage_fixation_nl: 0.9 ratio: 2.0
• estimator: tag: est_gnss_imu_camera_srr type: gnss_imu_camera_srr input_tags: [str_ros_camera, str_ros_imu, est_rtk] str_ros_camera_roles: [mono] str_ros_imu_roles: [major] est_rtk_roles: [position_and_velocity] output_tags: [fmt_srr_solution_file, str_solution, str_solution_odometry, str_solution_path, str_featured_image, str_landmarks] output_align_tag: str_ros_imu output_downsample_rate: 40
  compute_covariance: false enable_input_align: true input_align_latency: 0.2 enable_backend_data_sparsify: true
  pending_num_threshold: 5 gnss_imu_camera_srr_options: max_keyframes: 10
  min_yaw_std_init_visual: 5.0 gnss_loose_estimator_base_options: visual_estimator_base_options: feature_error_std: 10.0 stable_feature_error_std: 1.0 min_observation_stable: 20 landmark_outlier_rejection_threshold: 2.0 imu_estimator_base_options: imu_parameters: sigma_bg: 1.0e-3 sigma_ba: 1.0e-2 sigma_g_c: 2.6700e-04 sigma_a_c: 0.0112 sigma_gw_c: 2.0881e-06 sigma_aw_c: 2.7361e-04 body_to_imu_rotation: [-90.0, -90.0, 0.0]
  body_to_imu_rotation_std: 5.0 car_motion: true
  car_motion_min_velocity: 3.0 car_motion_max_anguler_velocity: 5.0 estimator_base_options:
  max_iteration: 5 num_threads: 4
  max_solver_time: 0.02 verbose_output: true solver_type: dense_schur trust_region_strategy_type: dogleg gnss_imu_initializer_options: max_iteration: 30 num_threads: 4 max_solver_time: 0.5 gnss_extrinsics: [-0.029, 0.354, -0.042] # gnss_extrinsics_initial_std: [0.0, 0.0, 0.0] time_window_length_slow_motion: 0.05 time_window_length_dynamic_motion: 0.5 min_acceleration: 0.5 feature_handler_options: max_features_per_frame: 100 max_n_kfs: 30 kfselect_min_numkfs: 30 kfselect_min_disparity: 10.0 kfselect_min_dist_metric: 0.5 kfselect_min_angle: 5.0
  kfselect_min_dt: 1.0 detector: cell_size: 40
  initialization: camera_parameters:
  label: "gici-board" cameras:
  • camera: label: cam0 image_width: 752 image_height: 480 type: pinhole intrinsics: cols: 1 rows: 4 data: [510.12091726559544, 509.55338413343736, 357.48596105253193, 240.00309579253124] distortion: type: radial-tangential
    parameters: cols: 1 rows: 4 data: [-0.36472323465957157, 0.11530151924890532, 0.0006024586154990503, 0.0020184970860498133] mask: /home/l1/gici-open/option/gici-mask.png T_B_C: cols: 4 rows: 4 data: [0.0134381210697455122, -0.999901594987671172, -0.00402706384669154413, -0.0229793554058591656, 0.999907681540912807, 0.0134460859659659704, -0.00195733688253543802, 0.0110786309679912626, 0.00201129251744842832, -0.00400038914436078377, 0.99998997574430859, 0.025008868367930974, 0, 0, 0, 1]

logging: enable: true min_log_level: 0 log_to_stderr: true file_directory: /home/l1/gici-open/output

[chichengcn]

你的配置文件看起来也没什么问题，这个问题我没有遇到过，你可以考虑根据程序执行顺序定位一下问题，以下是执行顺序：

（1）multisensorestimating.cpp - callback(tag, frame_data); （2）ros_stream.cpp - outputDataCallback - featuredImageOutputCallback(data->frame); （3）publishFeaturedImage(...)

我目前觉得有两种可能的情况：

1. multisensor_estimating.cpp那里执行callback之前，视觉前端的处理没有成功，导致输出的frame指针为空。如果这样的话，你可以检查一下feature_handler_相关的处理哪里出现了问题。
2. 由于内存问题导致异常调用。你可以监控一下上述执行路径中的一些变量，如tag、time，保证其值是合理的。

[936384885xy]

非常感谢您的回复，经过排查问题解决了。我的设备上有两个版本的opencv，自带的4.2版本opencv和我自己编译的4.5版本冲突了，删除4.2版本后就可以正常运行了。