The future of automated driving (AD) is rooted in the development of robust, fair and explainable artificial intelligence methods. Upon request, automated vehicles must be able to explain their decisions to the driver and the car passengers, to the pedestrians and other vulnerable road users and potentially to external auditors in case of accidents. However, nowadays, most explainable methods still rely on quantitative analysis of the AD scene representations captured by multiple sensors. This paper proposes a novel representation of AD scenes, called Qualitative eXplainable Graph (QXG), dedicated to qualitative spatiotemporal reasoning of long-term scenes. The construction of this graph exploits the recent Qualitative Constraint Acquisition paradigm. Our experimental results on NuScenes, an open real-world multi-modal dataset, show that the qualitative eXplainable graph of an AD scene composed of 40 frames can be computed in real-time and light in space storage which makes it a potentially interesting tool for improved and more trustworthy perception and control processes in AD.

De nombreux problèmes de planification et d’ordonnance-
ment impliquent la conception de subtiles combinaisons lo-
giques de contraintes temporelles ou spatiales. D’une part,
la modélisation précise de ces contraintes, qui sont formu-
lées dans diverses algèbres de relations, implique un cer-
tain nombre de combinaisons logiques possibles et néces-
site une expertise en modélisation basée sur les contraintes.
D’autre part, l’acquisition active de contraintes (AC) a été
utilisée avec succès pour aider les utilisateurs non expé-
rimentés à apprendre les réseaux de contraintes conjonc-
tives par la génération d’une séquence de requêtes. Dans
cet article, nous proposons GEQCA, pour Generic Quali-
tative Constraint Acquisition, une méthode d’AC active qui
apprend les contraintes qualitatives via le concept de re-
quêtes qualitatives. GEQCA combine les requêtes quali-
tatives avec la cohérence de chemin limitée dans le temps
(PC pour Path Consistency) et la propagation des connais-
sances de base pour acquérir les contraintes qualitatives.
Nous prouvons la correction, la complétude et la terminai-
son de GEQCA et nous donnons une évaluation expéri-
mentale qui montre (i) l’efficacité de notre approche dans
l’apprentissage des contraintes temporelles et, (ii) l’utilisa-
tion de GEQCA sur des instances réelles d’ordonnance-
ment.

Pattern mining is a valuable tool for exploratory data analysis, but identifying relevant patterns for a specific user is challenging. Various interestingness measures have been developed to evaluate patterns, but they may not efficiently estimate user-specific functions. Learning user-specific functions by ranking patterns has been proposed, but this requires significant time and training samples. In this paper, we present a solution that formulates the problem of learning pattern ranking functions as a multi-criteria decision-making problem. Our approach uses an analytic hierarchy process (AHP) to elicit weights for different interestingness measures based on user preference. We also propose an active learning mode with a sensitivity-based heuristic to minimize user ranking queries while still providing high-quality results. Experiments show that our approach significantly reduces running time and returns precise pattern ranking while being robust to user mistakes, compared to state-of-the-art approaches.

Many planning, scheduling or multi-dimensional packing problems involve the design of subtle logical combinations of temporal or spatial constraints. On the one hand, the pre- cise modelling of these constraints, which are formulated in various relation algebras, entails a number of possible logical combinations and requires expertise in constraint-based mod- elling. On the other hand, active constraint acquisition (CA) has been used successfully to support non-experienced users in learning conjunctive constraint networks through the gen- eration of a sequence of queries. In this paper, we propose GEQCA, which stands for Generic Qualitative Constraint Acquisition, an active CA method that learns qualitative con- straints via the concept of qualitative queries. GEQCA com- bines qualitative queries with time-bounded path consistency (PC) and background knowledge propagation to acquire the qualitative constraints of any scheduling or packing prob- lem. We prove soundness, completeness and termination of GEQCA by exploiting the jointly exhaustive and pairwise disjoint property of qualitative calculus and we give an ex- perimental evaluation that shows (i) the efficiency of our ap- proach in learning temporal constraints and, (ii) the use of GEQCA on real scheduling instances.