Motivation for ASAM OpenLABEL
From working with different customers, a significant fragmentation emerged in the way each individual organization categorizes and describes the objects populating the driving environment. Such categorizations and descriptions are the fundamental building block of any Autonomous Driving System’s (ADS) perception stack, since it is through them that an ADS come to a primal understanding of the status of around itself, including the entities present and some aspects of their behavior. Many vital driving decisions are based on this understanding.
The lack of a common Labeling in the industry is the root cause of several different issues: standard
- Hampered Vehicle2Vehicle Interaction: the different descriptions/understandings of surroundings may cause casualties in complex situations involving two or more different ADSs
- Precluded sharing: It results highly difficult if not impossible to share data across organizations adopting different Labeling taxonomies and specifications
- Lowered Annotation quality: Each individual labeling task requires ad-hoc training and even custom software features development to be completed, that translates into a higher probability of errors and thus a threat to safety
- Deprecation of old labels: Long-term operation of ADS development imply changes in quantity and richness of labels to be produced, considering the evolution of the driving scenes, new sensors, and scenarios. As a consequence, a flexible descriptive language is required to absorb future extensions/modifications of labels and guarantee back-compatibility.
In sum, the absence of a labeling such as OpenLABEL is ultimately a significant safety threat for all road users surrounding any kind of vehicle which is being operated in autonomous or semi- autonomous ( standard Level >=2) mode. OpenLABEL objective is to increase overall operational safety by providing a language that allows for the encoding of a common baseline understanding of the driving environment for any ADS. SAE