Assessing the traversability of rugged terrain is a difficult challenge for legged robots, especially when they implement multiple, distinct gaits. We tackle this problem on the k-rock2 amphibious, sprawling gait robot by training a gait-dependent traversability estimator. We verify that the estimator, trained solely on procedurally-generated simulated data, approaches the outcomes of real-world experiments conducted in an indoor motion capture arena using two distinct terrestrial gaits to cross various indoor obstacles. In simulation experiments on a large-scale outdoor heightmap representing real-world data, we quantify the performance gain using the estimator outputs for gait selection. Further, we apply the method to heightmaps of outdoor data to illustrate how the approach could readily be applied to field scenarios.

We demonstrate a self-supervised approach which learns to detect long-range obstacles from video: it automatically obtains training labels by associating the camera frames acquired at a given pose to short-range sensor readings acquired at a different pose.

We consider the problem of planning paths on graphs with some edges whose traversability is uncertain; for each uncertain edge, we are given a probability of being traversable (e.g., by a learned classifier). We categorize different interpretations of the problem that are meaningful for mobile robots navigating partially-known environments, each of which yields a different formalization; we then focus on the case in which the true traversability of an edge is revealed only when the agent visits one of its endpoints (Canadian Traveller Problem). In this context, we design a large simulation campaign on synthetic and real-world maps to study the impact of two different factors: the planning strategy, and the amount of uncertainty (which could depend on the quality of the classifier producing traversability estimates).

Considering perception as an observation process only is the very reason for which robotic perception methods are to date unable to provide a general capacity of scene understanding. Related work in neuroscience has shown that there is a strong relationship between perception and action. We believe that considering perception in relation to action requires to interpret the scene in terms of the agent's own potential capabilities. In this paper, we propose a Bayesian approach for learning sensorimotor representations through the interaction between action and observation capabilities. We represent the notion of affordance as a probabilistic relation between three elements: objects, actions and effects. Experiments for affordances discovery were performed on a real robotic platform in an unsupervised way assuming a limited set of innate capabilities. Results show dependency relations that connect the three elements in a common frame: affordances. The increasing number of interactions and observations results in a Bayesian network that captures the relationships between them. The learned representation can be used for prediction tasks.

The accurate detection and classification of moving objects is a critical aspect of advanced driver assistance systems. We believe that by including the object classification from multiple sensor detections as a key component of the object's representation and the perception process, we can improve the perceived model of the environment. First, we define a composite object representation to include class information in the core object's description. Second, we propose a complete perception fusion architecture based on the evidential framework to solve the detection and tracking of moving objects problem by integrating the composite representation and uncertainty management. Finally, we integrate our fusion approach in a real-time application inside a vehicle demonstrator from the interactIVe IP European project, which includes three main sensors: radar, lidar, and camera. We test our fusion approach using real data from different driving scenarios and focusing on four objects of interest: pedestrian, bike, car, and truck.

Intelligent vehicle perception involves the correct detection and tracking of moving objects. Taking into account all the possible information at early levels of the perception task can improve the final model of the environment. In this paper, we present an evidential fusion framework to represent and combine evidence from multiple lists of sensor detections. Our fusion framework considers the position, shape and appearance information to represent, associate and combine sensor detections. Although our approach takes place at detection level, we propose a general architecture to include it as a part of a whole perception solution. Several experiments were conducted using real data from a vehicle demonstrator equipped with three main sensors: lidar, radar and camera. The obtained results show improvements regarding the reduction of false detections and mis-classifications of moving objects.

In this paper, we detail a complete software architecture of a key task that an intelligent vehicle has to deal with: frontal object perception. This task is solved by processing raw data of a radar and a mono-camera to detect and track moving objects. Data sets obtained from highways, country roads and urban areas were used to test the proposed method. Several experiments were conducted to show that the proposed method obtains a better environment representation, i.e., reduces the false alarms and missed detections from individual sensor evidence.