Vehicle obstacle avoidance is not available

[1111F11]

I deployed autoware.universe to a car during an experiment. The car can now drive autonomously but cannot avoid obstacles. Where should I look for the problem? Please help me.
I looked at the link diagram and did the plaining part of the query as follows：

autoware@autoware:~$ ros2 topic echo /planning/scenario_planning/trajectory
header:
stamp:
sec: 1725329384
nanosec: 36276761
frame_id: map
points:

• time_from_start:
  sec: 0
  nanosec: 0
  pose:
  position:
  x: -11.386725579632056
  y: -46.79024569654417
  z: -1.3743683746410875
  orientation:
  x: 0.006003021982939154
  y: 0.005973249987614024
  z: -0.7088369794699565
  w: 0.7053215015491633
  longitudinal_velocity_mps: 0.25
  lateral_velocity_mps: 0.0
  acceleration_mps2: 0.10000000149011612
  heading_rate_rps: 0.0
  front_wheel_angle_rad: 0.0
  rear_wheel_angle_rad: 0.0

autoware@autoware:~$ ros2 topic echo /perception/object_recognition/objects
header:
stamp:
sec: 1725329432
nanosec: 296296768
frame_id: map
objects: []

autoware@autoware:~$ ros2 node info /planning/scenario_planning/lane_driving/motion_planning/obstacle_stop_planner
/planning/scenario_planning/lane_driving/motion_planning/obstacle_stop_planner
Subscribers:
/localization/acceleration: geometry_msgs/msg/AccelWithCovarianceStamped
/localization/kinematic_state: nav_msgs/msg/Odometry
/parameter_events: rcl_interfaces/msg/ParameterEvent
/perception/object_recognition/objects: autoware_auto_perception_msgs/msg/PredictedObjects
/perception/obstacle_segmentation/pointcloud: sensor_msgs/msg/PointCloud2
/planning/scenario_planning/lane_driving/motion_planning/obstacle_stop_planner/input/expand_stop_range: tier4_planning_msgs/msg/ExpandStopRange
/planning/scenario_planning/lane_driving/motion_planning/obstacle_velocity_limiter/trajectory: autoware_auto_planning_msgs/msg/Trajectory
Publishers:
/parameter_events: rcl_interfaces/msg/ParameterEvent
/planning/scenario_planning/clear_velocity_limit: tier4_planning_msgs/msg/VelocityLimitClearCommand
/planning/scenario_planning/lane_driving/motion_planning/obstacle_stop_planner/adaptive_cruise_control/debug_values: tier4_debug_msgs/msg/Float32MultiArrayStamped
/planning/scenario_planning/lane_driving/motion_planning/obstacle_stop_planner/debug/collision_pointcloud: sensor_msgs/msg/PointCloud2
/planning/scenario_planning/lane_driving/motion_planning/obstacle_stop_planner/debug/marker: visualization_msgs/msg/MarkerArray
/planning/scenario_planning/lane_driving/motion_planning/obstacle_stop_planner/debug/obstacle_pointcloud: sensor_msgs/msg/PointCloud2
/planning/scenario_planning/lane_driving/motion_planning/obstacle_stop_planner/debug/processing_time_ms: tier4_debug_msgs/msg/Float64Stamped
/planning/scenario_planning/lane_driving/motion_planning/obstacle_stop_planner/obstacle_stop/debug_values: tier4_debug_msgs/msg/Float32MultiArrayStamped
/planning/scenario_planning/lane_driving/motion_planning/obstacle_stop_planner/output/stop_reason: diagnostic_msgs/msg/DiagnosticStatus
/planning/scenario_planning/lane_driving/motion_planning/obstacle_stop_planner/virtual_wall: visualization_msgs/msg/MarkerArray
/planning/scenario_planning/lane_driving/trajectory: autoware_auto_planning_msgs/msg/Trajectory
/planning/scenario_planning/max_velocity_candidates: tier4_planning_msgs/msg/VelocityLimit
/planning/scenario_planning/status/stop_reasons: tier4_planning_msgs/msg/StopReasonArray
/planning/velocity_factors/obstacle_stop: autoware_adapi_v1_msgs/msg/VelocityFactorArray
/rosout: rcl_interfaces/msg/Log
Service Servers:
/planning/scenario_planning/lane_driving/motion_planning/obstacle_stop_planner/config_logger: logging_demo/srv/ConfigLogger
/planning/scenario_planning/lane_driving/motion_planning/obstacle_stop_planner/describe_parameters: rcl_interfaces/srv/DescribeParameters
/planning/scenario_planning/lane_driving/motion_planning/obstacle_stop_planner/get_parameter_types: rcl_interfaces/srv/GetParameterTypes
/planning/scenario_planning/lane_driving/motion_planning/obstacle_stop_planner/get_parameters: rcl_interfaces/srv/GetParameters
/planning/scenario_planning/lane_driving/motion_planning/obstacle_stop_planner/list_parameters: rcl_interfaces/srv/ListParameters
/planning/scenario_planning/lane_driving/motion_planning/obstacle_stop_planner/set_parameters: rcl_interfaces/srv/SetParameters
/planning/scenario_planning/lane_driving/motion_planning/obstacle_stop_planner/set_parameters_atomically: rcl_interfaces/srv/SetParametersAtomically
Service Clients:

Action Servers:

Action Clients:

[maxime-clem]

autoware@autoware:~$ ros2 topic echo /perception/object_recognition/objects
objects: []

If objects are not detected then obstacle avoidance will no work.
Can you first confirm if perception works correctly ?

I think the only "avoidance" that can be done without object detection is with AEB (https://autowarefoundation.github.io/autoware.universe/latest/control/autoware_autonomous_emergency_braking/) using pointcloud data.

[1111F11]

I think there is no problem with the perception layer. Today’s cars can implement behaviors such as path planning and turning, but they cannot identify obstacles like in the simulation.

What parameters should I adjust?

[maxime-clem]

The 2nd picture is recorded on a real vehicle right ?
It looks like the raw pointcloud (red) is incorrectly filtered and the resulting "no ground" pointcloud (white) does not contain the obstacles.
You may need to calibrate your LIDAR and you can find a guide here: https://autowarefoundation.github.io/autoware-documentation/main/how-to-guides/integrating-autoware/creating-vehicle-and-sensor-model/calibrating-sensors/ground-lidar-calibration/