ODS (Obstacle Detection System)

ODS - Obstacle Detection System - provided by ifm, is a software solution building on top of the O3R ecosystem to enable autonomously driving vehicles (AGVs / AMRs) to avoid collisions with obstacles.

The ODS application is specifically designed to detect two groups of objects, which are missed by a standard LiDAR based autonomously driving vehicle:

  • objects above the safety LiDAR scanner plane, that is, cantilevered objects,

  • and objects below the safety LiDAR scanner plane, that is, small sized objects on the floor.

Its application lies within the addition of a full 3D volumetric monitoring of the surrounding of autonomously driving vehicles - whereas LiDAR scanner planes are limited to two spatial dimensions. ODS doesn’t follow any safety regulations: for example ASIL-level or similar, and can therefore not replace a typical application of safety devices such as safety LiDAR scanners.

ODS application properties and building blocks:

  • ODS uses the O3R cameras as its primary data source: at least one O3R 3D camera stream is required

  • ODS is capable of estimating the required ego-motion information via an internal algorithm - that is, is not dependent on external ego-motion sources

  • ODS outputs two kind of information sets:

    • A 2D occupancy grid information: typically used in addition to existing navigation information to avoid and navigate around obstacles

    • (3D) zone information: a typical use case is the direct propagation to the chassis control system to perform braking maneuvers when obstacles are present

  • ODS does not classify objects: the output information is independent of the object type

  • multiple ODS applications can be initialized on one O3R VPU (IPC): one application instance per primary movement direction - that is, forward, backward driving

  • ODS is a licensed application: Please see the O3Rs release notes for additional information for compatibility tables and embedded OS versions