BiguaSim Sensors
BiguaSim sensors are used to gather data from the environment. Sensors are attached to agents at
specific locations called “sockets” (see BiguaSim Agents for socket locations on each agent). Sensors
publish data to the simulation state, which can be accessed as the return value of the environment’s
.tick() and .step() functions.
Using Sensors
BiguaSim Sensors
The following are inspired by real-world sensors and are intended to collect realistic data. Most sensors here include parameters for adjusting noise, bias, and other characteristics to simulate real-world sensor behavior.
Perception Sensors
- Single-Beam Sonar
- Sidescan Sonar
- Imaging Sonar
- Profiling Sonar
- Doppler Velocity Log (DVL)
- Depth Sensor
- RGB Camera
- Camera
- Inertial Measurement Unit (IMU)
- Global Positioning System (GPS)
- Magnetometer
- Range Finder Sensor
- RGBD (Depth) Camera
- Semantic Secmentation Camera
- Raycast LIDAR
- Semantic Segmentation Raycast Lidar
Communication Sensors
State Sensors
The following are not physical sensors, but are used to return information that is only available in simulation. They are helpful in debugging, collecting ground truth data, capturing simulation videos, and other applications where simulation information is helpful. Some of these have overlap with each other, but are provided for convenience for different use cases.
Note
The location, rotation, orientation, pose, and velocity sensors return values relative to the global frame for the pose of the socket they are placed in.
The dynamics sensor always returns values for the COM of the agent, with values given in the global frame, regardless of the socket it is placed in.