数据说明
话题目录如下:
/baton/CamL2Imu
/baton/CamR2Imu
/baton/algo_status
/baton/build_map
/baton/camera_left_info
/baton/camera_right_info
/baton/depth_test_image
/baton/depth_version
/baton/image_left
/baton/image_left/compressed
/baton/image_left/compressedDepth
/baton/image_left/theora
/baton/image_left_info
/baton/image_right
/baton/image_right/compressed
/baton/image_right/compressedDepth
/baton/image_right/theora
/baton/image_right_info
/baton/imu
/baton/imu_temp
/baton/stereo3/feature_img
/baton/stereo3/feature_img/compressed
/baton/stereo3/feature_img/compressedDepth
/baton/stereo3/feature_img/theora
/baton/stereo3/fusion_odom
/baton/stereo3/fusion_path
/baton/stereo3/imu_odom
/baton/stereo3/map_odom
/baton/stereo3/map_path
/baton/stereo3/odom_path
/baton/stereo3/odom_relo
/baton/stereo3/odometry
/baton/stereo3/pointcloud
/baton/stereo3_ctrl
/baton/stereo4/feature_img
/baton/stereo4/feature_img/compressed
/baton/stereo4/feature_img/compressedDepth
/baton/stereo4/feature_img/theora
/baton/stereo4/map_odom
/baton/stereo4/map_path
/baton/stereo4/odom_path
/baton/stereo4/odom_relo
/baton/stereo4/odometry
/baton/stereo4_ctrl
/parameter_events
/rosout
一.输出话题
输出话题分为传感器数据和算法数据两类。
1.传感器数据输出
(1)左目图像
话题名:/baton/image_left
Type: sensor_msgs/Image
(2)右目图像
话题名:/baton/image_right
Type: sensor_msgs/Image
(3)左目图像曝光信息
话题名:/baton/image_left_info
Type: sensor_pub/ImageInfo
该消息类型为自定义类型,在自定义消息类型可以找到,具体定义如下。
exposure: 4976 #曝光时间
analog_gain: 1.0 #模拟增益
digital_gain: 1.0 #数字增益
(4)右目图像曝光信息
话题名:/baton/image_right_info
Type: sensor_pub/ImageInfo
该消息类型为自定义类型,定义如上。
(5)imu原始数据
话题名:/baton/imu
Type: sensor_msgs/Imu
(6)imu温度数据
话题名:/baton/imu_temp
Type: sensor_msgs/Temperature
(7)左目到IMU外参
话题名:/baton/CamL2Imu
Type: geometry_msgs/PoseStamped
(8)右目到IMU外参
话题名:/baton/CamR2Imu
Type: geometry_msgs/PoseStamped
(9)左目参数
话题名:/baton/camera_left_info
Type: sensor_msgs/CameraInfo
height: 480 #图像高
width: 640 #图像宽
distortion_model: "ds" #畸变模型
D: [-0.25824591334560476, 0.5672715939661027, 0.0, 0.0, 0.0] #畸变系数
K: [164.15676583809704, 0.0, 318.5936994457823,
0.0, 164.0741011283019, 239.66298170413694,
0.0, 0.0, 1.0]#相机内参
R: [1.0, 0.0, 0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 1.0]
P: [1.0, 0.0, 0.0, 0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 1.0, 0.0]
binning_x: 0
binning_y: 0
roi:
x_offset: 0
y_offset: 0
height: 0
width: 0
do_rectify: False
(10)右目参数
话题名:/baton/camera_right_info
Type: sensor_msgs/CameraInfo
height: 480 #图像高
width: 640 #图像宽
distortion_model: "ds" #畸变模型
D: [-0.2615917683666564, 0.5679955819508452, 0.0, 0.0, 0.0] #畸变系数
K: [163.79787865873183, 0.0, 323.4330173880752,
0.0, 163.62052609635333, 243.67927518142795,
0.0, 0.0, 1.0]#相机内参
R: [0.9999730625093293, 0.0031151927461599226, 0.006646038659791247,
-0.0031351685298500703, 0.9999905932027023, 0.002997369579618926,
-0.006636638757880752, -0.003018125289258183, 0.9999734226196892]#右目到左目外参旋转矩阵
P: [ 163.79787865873183, 0.0, 323.4330173880752,0.061327748008980894,
0.0, 163.62052609635333, 243.67927518142795, -0.0004044248281003861,
0.0, 0.0, 1.0,-0.00033629396372312104] #最后一列为右目到左目外参平移变换
binning_x: 0
binning_y: 0
roi:
x_offset: 0
y_offset: 0
height: 0
width: 0
do_rectify: False
2.算法输出
(1)算法状态
话题名:/baton/algo_status
Type:system_ctrl/algo_status
该话题为自定义话题消息类型,在自定义消息里面能找到,具体定义如下
algo_status: "ready" #或者是“stereo3_inilizing”或"stereo3_running"
(2)特征点图
话题名:/baton/stereo3/feature_img
Type:sensor_msgs/Image
算法正常启动后,通过左右目提取特征点并跟踪,合成在同一张图片上显示出跟踪到的特征点。
(3)stereo3输出里程计
话题名:/baton/stereo3/odometry
Type:nav_msgs/Odometry
算法正常启动后,通过VIO计算出来的当前位姿与开始第一帧位姿(0,0,0,0,0,0,0)之间的相对位姿。
(4)stereo3输出点云
话题名:/baton/stereo3/pointcloud
Type:sensor_msgs/PointCloud2
算法正常启动后,通过VIO计算特征点投影到三维空间中生成的点云。相对来说会比较稀疏,而且呈散射状。
(5)stereo3运行轨迹
话题名:/baton/stereo3/path_odom
Type:nav_msgs/Path
算法正常启动后,通过stereo3运行过程中的历史位姿连成的轨迹。
(6)回环位姿
话题名:/baton/loop/odometry
Type:nav_msgs/``Odometry
算法正常启动后,通过回环修正后的位姿,如果没有触发过回环,它的数据应该是与/baton/stereo3/odometry一致的。
(7)回环点云
话题名:/baton/loop/pointcloud
Type:sensor_msgs/PointCloud2
算法正常启动后,通过修正过后的点云。
(8)回环轨迹
话题名:/baton/loop/path_odom
Type:nav_msgs``/``Path
算法正常启动后,通过修正过后的轨迹。
二.控制话题
1.算法控制
(1)stereo3控制
话题名:/baton/stereo3_ctrl
Type:system_ctrl/algo_ctrl
用于控制stereo3算法的启停、重启。该话题为自定义话题,在自定义消息类型可以找到,具体定义如下:
Header header
bool algo_enable
bool algo_reboot
bool algo_reset
(2)回环控制
主要是控制关键帧的保存的,action服务器为/baton/loop/keyframe_action/
对应ROS的action使用即可。