Search

Accurate and robust trajectory predictions of road users are needed to enable safe automated driving. To do this, machine learning models are often used, which can show erratic behavior when presented with previously unseen inputs. In this work, two environment-aware models (MotionCNN and MultiPath++) and two common baselines (Constant Velocity and an LSTM) are benchmarked for robustness against various perturbations that simulate functional insufficiencies observed during model deployment in a vehicle: unavailability of road information, late detections, and noise. Results show significant performance degradation under the presence of these perturbations, with errors increasing up to +1444.8\% in commonly used trajectory prediction evaluation metrics. Training the models with similar perturbations effectively reduces performance degradation, with error increases of up to +87.5\%. We argue that despite being an effective mitigation strategy, data augmentation through perturbations during training does not guarantee robustness towards unforeseen perturbations, since identification of all possible on-road complications is unfeasible. Furthermore, degrading the inputs sometimes leads to more accurate predictions, suggesting that the models are unable to learn the true relationships between the different elements in the data.

An automated vehicle (AV) must always have a correct representation of the drivable space to position itself accurately and operate safely. To determine the drivable space, current research focuses on single sources of information, either using pre-computed high-definition maps, or mapping the environment online with sensors such as LiDARs or cameras. However, each of these information sources can fail, some are too costly, and maps could be outdated. In this work a new method for situation-aware drivable space (SDS) estimation combining multiple information sources is proposed, which is also suitable for AVs equipped with inexpensive sensors. Depending on the situation, semantic information of sensed objects is combined with domain knowledge to estimate the drivability of the space surrounding each object (e.g. traffic light, another vehicle). These estimates are modeled as probabilistic graphs to account for the uncertainty of information sources, and an optimal spatial configuration of their elements is determined via graph-based simultaneous localization and mapping (SLAM). To investigate the robustness of SDS towards potentially unreliable sensors and maps, it has been tested in a simulation environment and real world data. Results on different use cases (e.g. straight roads, curved roads, and intersections) show considerable robustness towards unreliable inputs, and the recovered drivable space allows for accurate in-lane localization of the AV even in extreme cases where no prior knowledge of the road network is available.

To operate safely, an automated vehicle (AV) must anticipate how the environment around it will evolve. For that purpose, it is important to know which prediction models are most appropriate for every situation. Currently, assessment of prediction models is often performed over a set of trajectories without distinction of the type of movement they capture, resulting in the inability to determine the suitability of each model for different situations. In this work we illustrate how standardized evaluation methods result in wrong conclusions regarding a model's predictive capabilities, preventing a clear assessment of prediction models and potentially leading to dangerous on-road situations. We argue that following evaluation practices in safety assessment for AVs, assessment of prediction models should be performed in a scenario-based fashion. To encourage scenario-based assessment of prediction models and illustrate the dangers of improper assessment, we categorize trajectories of the Waymo Open Motion dataset according to the type of movement they capture. Next, three different models are thoroughly evaluated for different trajectory types and prediction horizons. Results show that common evaluation methods are insufficient and the assessment should be performed depending on the application in which the model will operate.

This paper presents an automated flower counting method based on Multiple Hypothesis Tracking (MHT) with a connected-flower plant model which is based on detections of flowers. Multiple viewpoints of each plant are taken into account as plants are considered in which flowers can occlude each other. To prevent double counting and to solve inconsistencies caused by false flower detections, a model is developed which describes the plant movement with respect to the camera. The uncertainty of the flower detections is considered in this model. To address variations in the velocity of the plant movement, the model realized in this work explicitly takes into account that motions of flowers are correlated since the flowers are connected to each other via the stem of the plant. This is in contrast to the traditional MHT approach where the movement of each object is typically modeled and estimated separately. In our approach, based on the set of detected flowers, the uncertainty of the plant movement is reduced. As a result, the movement of modeled but not always observed flowers is still properly tracked. To demonstrate the validity of the approach, the proposed counting method is tested on a dataset obtained in a real greenhouse containing multiple viewpoints of 71 Phalaenopsis plants and compared to existing methods. The methods considered include a single viewpoint approach, a heuristic state of the practice approach and an MHT approach with both an independent and connected object description. Within a margin of 1 flower, these methods respectively counted the number of flowers in 44%,58%,70% and 92% of the plants correctly. As a result, this work validates the superiority of the MHT approach with a connected-flower plant model.

In this work, we propose and evaluate a pose-graph optimization-based real-time multi-sensor fusion framework for vehicle positioning using low-cost automotive-grade sensors. Pose-graphs can model multiple absolute and relative vehicle positioning sensor measurements and can be optimized using nonlinear techniques. We model pose-graphs using measurements from a precise stereo camera-based visual odometry system, a robust odometry system using the in-vehicle velocity and yaw-rate sensor, and an automotive-grade GNSS receiver. Our evaluation is based on a dataset with 180 km of vehicle trajectories recorded in highway, urban, and rural areas, accompanied by postprocessed Real-Time Kinematic GNSS as ground truth. We compare the architecture’s performance with (i) vehicle odometry and GNSS fusion and (ii) stereo visual odometry, vehicle odometry, and GNSS fusion; for offline and real-time optimization strategies. The results exhibit a 20.86% reduction in the localization error’s standard deviation and a significant reduction in outliers when compared with automotive-grade GNSS receivers.