Ray Caster

Ray Caster#

A diagram outlining the basic geometry of frame transformations

The Ray Caster sensor (and the ray caster camera) are similar to RTX based rendering in that they both involve casting rays. The difference here is that the rays cast by the Ray Caster sensor return strictly collision information along the cast, and the direction of each individual ray can be specified. They do not bounce, nor are they affected by things like materials or opacity. For each ray specified by the sensor, a line is traced along the path of the ray and the location of first collision with the specified mesh is returned. This is the method used by some of our quadruped examples to measure the local height field.

To keep the sensor performant when there are many cloned environments, the line tracing is done directly in Warp. This is the reason why specific meshes need to be identified to cast against: that mesh data is loaded onto the device by warp when the sensor is initialized. As a consequence, the current iteration of this sensor only works for literally static meshes (meshes that are not changed from the defaults specified in their USD file). This constraint will be removed in future releases.

Using a ray caster sensor requires a pattern and a parent xform to be attached to. The pattern defines how the rays are cast, while the prim properties defines the orientation and position of the sensor (additional offsets can be specified for more exact placement). Isaac Lab supports a number of ray casting pattern configurations, including a generic LIDAR and grid pattern.



@configclass
class RaycasterSensorSceneCfg(InteractiveSceneCfg):
    """Design the scene with sensors on the robot."""

    # ground plane
    ground = AssetBaseCfg(
        prim_path="/World/Ground",
        spawn=sim_utils.UsdFileCfg(
            usd_path=f"{ISAAC_NUCLEUS_DIR}/Environments/Terrains/rough_plane.usd",
            scale=(1, 1, 1),
        ),
    )

    # lights
    dome_light = AssetBaseCfg(
        prim_path="/World/Light", spawn=sim_utils.DomeLightCfg(intensity=3000.0, color=(0.75, 0.75, 0.75))
    )

    # robot
    robot = ANYMAL_C_CFG.replace(prim_path="{ENV_REGEX_NS}/Robot")

    ray_caster = RayCasterCfg(
        prim_path="{ENV_REGEX_NS}/Robot/base",
        update_period=1 / 60,
        offset=RayCasterCfg.OffsetCfg(pos=(0, 0, 0.5)),
        mesh_prim_paths=["/World/Ground"],
        ray_alignment="yaw",
        pattern_cfg=patterns.LidarPatternCfg(
            channels=100, vertical_fov_range=[-90, 90], horizontal_fov_range=[-90, 90], horizontal_res=1.0
        ),

Notice that the units on the pattern config is in degrees! Also, we enable visualization here to explicitly show the pattern in the rendering, but this is not required and should be disabled for performance tuning.

Lidar Pattern visualized

Querying the sensor for data can be done at simulation run time like any other sensor.

def run_simulator(sim: sim_utils.SimulationContext, scene: InteractiveScene):
  .
  .
  .
  # Simulate physics
  while simulation_app.is_running():
    .
    .
    .
    # print information from the sensors
      print("-------------------------------")
      print(scene["ray_caster"])
      print("Ray cast hit results: ", scene["ray_caster"].data.ray_hits_w)
-------------------------------
Ray-caster @ '/World/envs/env_.*/Robot/base/lidar_cage':
        view type            : <class 'isaacsim.core.experimental.prims.xform_prim.XformPrim'>
        update period (s)    : 0.016666666666666666
        number of meshes     : 1
        number of sensors    : 1
        number of rays/sensor: 18000
        total number of rays : 18000
Ray cast hit results:  tensor([[[-0.3698,  0.0357,  0.0000],
        [-0.3698,  0.0357,  0.0000],
        [-0.3698,  0.0357,  0.0000],
        ...,
        [    inf,     inf,     inf],
        [    inf,     inf,     inf],
        [    inf,     inf,     inf]]], device='cuda:0')
-------------------------------

Here we can see the data returned by the sensor itself. Notice first that there are 3 closed brackets at the beginning and the end: this is because the data returned is batched by the number of sensors. The ray cast pattern itself has also been flattened, and so the dimensions of the array are [N, B, 3] where N is the number of sensors, B is the number of cast rays in the pattern, and 3 is the dimension of the casting space. Finally, notice that the first several values in this casting pattern are the same: this is because the lidar pattern is spherical and we have specified our FOV to be hemispherical, which includes the poles. In this configuration, the “flattening pattern” becomes apparent: the first 180 entries will be the same because it’s the bottom pole of this hemisphere, and there will be 180 of them because our horizontal FOV is 180 degrees with a resolution of 1 degree.

You can use this script to experiment with pattern configurations and build an intuition about how the data is stored by altering the triggered variable on line 81.

Code for raycaster_sensor.py
  1# Copyright (c) 2022-2026, The Isaac Lab Project Developers (https://github.com/isaac-sim/IsaacLab/blob/main/CONTRIBUTORS.md).
  2# All rights reserved.
  3#
  4# SPDX-License-Identifier: BSD-3-Clause
  5
  6import argparse
  7
  8from isaaclab.app import AppLauncher
  9
 10# add argparse arguments
 11parser = argparse.ArgumentParser(description="Example on using the raycaster sensor.")
 12parser.add_argument("--num_envs", type=int, default=1, help="Number of environments to spawn.")
 13# append AppLauncher cli args
 14AppLauncher.add_app_launcher_args(parser)
 15# demos should open Kit visualizer by default
 16parser.set_defaults(visualizer=["kit"])
 17# parse the arguments
 18args_cli = parser.parse_args()
 19
 20# launch omniverse app
 21app_launcher = AppLauncher(args_cli)
 22simulation_app = app_launcher.app
 23
 24"""Rest everything follows."""
 25
 26import numpy as np
 27import torch
 28
 29import isaaclab.sim as sim_utils
 30from isaaclab.assets import AssetBaseCfg
 31from isaaclab.scene import InteractiveScene, InteractiveSceneCfg
 32from isaaclab.sensors.ray_caster import RayCasterCfg, patterns
 33from isaaclab.utils import configclass
 34from isaaclab.utils.assets import ISAAC_NUCLEUS_DIR
 35
 36##
 37# Pre-defined configs
 38##
 39from isaaclab_assets.robots.anymal import ANYMAL_C_CFG  # isort: skip
 40
 41
 42@configclass
 43class RaycasterSensorSceneCfg(InteractiveSceneCfg):
 44    """Design the scene with sensors on the robot."""
 45
 46    # ground plane
 47    ground = AssetBaseCfg(
 48        prim_path="/World/Ground",
 49        spawn=sim_utils.UsdFileCfg(
 50            usd_path=f"{ISAAC_NUCLEUS_DIR}/Environments/Terrains/rough_plane.usd",
 51            scale=(1, 1, 1),
 52        ),
 53    )
 54
 55    # lights
 56    dome_light = AssetBaseCfg(
 57        prim_path="/World/Light", spawn=sim_utils.DomeLightCfg(intensity=3000.0, color=(0.75, 0.75, 0.75))
 58    )
 59
 60    # robot
 61    robot = ANYMAL_C_CFG.replace(prim_path="{ENV_REGEX_NS}/Robot")
 62
 63    ray_caster = RayCasterCfg(
 64        prim_path="{ENV_REGEX_NS}/Robot/base",
 65        update_period=1 / 60,
 66        offset=RayCasterCfg.OffsetCfg(pos=(0, 0, 0.5)),
 67        mesh_prim_paths=["/World/Ground"],
 68        ray_alignment="yaw",
 69        pattern_cfg=patterns.LidarPatternCfg(
 70            channels=100, vertical_fov_range=[-90, 90], horizontal_fov_range=[-90, 90], horizontal_res=1.0
 71        ),
 72        debug_vis=not args_cli.headless,
 73    )
 74
 75
 76def run_simulator(sim: sim_utils.SimulationContext, scene: InteractiveScene):
 77    """Run the simulator."""
 78    # Define simulation stepping
 79    sim_dt = sim.get_physics_dt()
 80    sim_time = 0.0
 81    count = 0
 82
 83    triggered = True
 84    countdown = 42
 85
 86    # Simulate physics
 87    while simulation_app.is_running():
 88        if count % 500 == 0:
 89            # reset counter
 90            count = 0
 91            # reset the scene entities
 92            # root state
 93            # we offset the root state by the origin since the states are written in simulation world frame
 94            # if this is not done, then the robots will be spawned at the (0, 0, 0) of the simulation world
 95            root_pose = scene["robot"].data.default_root_pose.torch.clone()
 96            root_pose[:, :3] += scene.env_origins
 97            scene["robot"].write_root_pose_to_sim_index(root_pose=root_pose)
 98            root_vel = scene["robot"].data.default_root_vel.torch.clone()
 99            scene["robot"].write_root_velocity_to_sim_index(root_velocity=root_vel)
100            # set joint positions with some noise
101            joint_pos, joint_vel = (
102                scene["robot"].data.default_joint_pos.torch.clone(),
103                scene["robot"].data.default_joint_vel.torch.clone(),
104            )
105            joint_pos += torch.rand_like(joint_pos) * 0.1
106            scene["robot"].write_joint_position_to_sim_index(position=joint_pos)
107            scene["robot"].write_joint_velocity_to_sim_index(velocity=joint_vel)
108            # clear internal buffers
109            scene.reset()
110            print("[INFO]: Resetting robot state...")
111        # Apply default actions to the robot
112        # -- generate actions/commands
113        targets = scene["robot"].data.default_joint_pos.torch
114        # -- apply action to the robot
115        scene["robot"].set_joint_position_target_index(target=targets)
116        # -- write data to sim
117        scene.write_data_to_sim()
118        # perform step
119        sim.step()
120        # update sim-time
121        sim_time += sim_dt
122        count += 1
123        # update buffers
124        scene.update(sim_dt)
125
126        # print information from the sensors
127        print("-------------------------------")
128        print(scene["ray_caster"])
129        print("Ray cast hit results: ", scene["ray_caster"].data.ray_hits_w.torch)
130
131        if not triggered:
132            if countdown > 0:
133                countdown -= 1
134                continue
135            data = scene["ray_caster"].data.ray_hits_w.torch.cpu().numpy()
136            np.save("cast_data.npy", data)
137            triggered = True
138        else:
139            continue
140
141
142def main():
143    """Main function."""
144
145    # Initialize the simulation context
146    sim_cfg = sim_utils.SimulationCfg(dt=0.005, device=args_cli.device)
147    sim = sim_utils.SimulationContext(sim_cfg)
148    # Set main camera
149    sim.set_camera_view(eye=[3.5, 3.5, 3.5], target=[0.0, 0.0, 0.0])
150    # design scene
151    scene_cfg = RaycasterSensorSceneCfg(num_envs=args_cli.num_envs, env_spacing=2.0)
152    scene = InteractiveScene(scene_cfg)
153    # Play the simulator
154    sim.reset()
155    # Now we are ready!
156    print("[INFO]: Setup complete...")
157    # Run the simulator
158    run_simulator(sim, scene)
159
160
161if __name__ == "__main__":
162    # run the main function
163    main()
164    # close sim app
165    simulation_app.close()