Road surface condition (RSC) estimation is useful to warn a driver or for automatic speed control in slippery road conditions. By using front-facing camera images, the RSC few meters ahead of the vehicle can be estimated, which could provide the valuable milliseconds to act in critical conditions before reaching the slippery road region. Convolutional neural networks (CNNs) have been used in previous work to classify or segment the RSC from front-facing camera images. In this work, we look for ways in improving the performance for RSC classification by fusion additional information. In-contrast to the widely used approach of classifying the most prominent RSC of the whole image, we propose to separate the road region into a collection of cells using a 2D grids and classifying the RSC of each grid-cell into dry-moist, wet-water, slush, snow-ice classes. Three additional information sources are investigated to fuse with road grid-cell region of the image, which are from the temperature sensor and other areas of the image, in an alternative manner. Our results indicates clear improvements on RSC estimation performance when fusing with each of the three sources of additional information, compared to only using image grid-cell region. The best performance with a 5.8% increase in average F1-score is obtained when using the full image in addition to the grid-cell image region.

This paper studies the optimization of ride comfort during the maneuver of bus docking at a stop station. We propose an analytical comfort model that considers the coupled and nonlinear effect of acceleration and jerk levels on the comfort perceived by standing bus passengers. This is studied through offline path planning by formulating the docking problem as a Nonlinear Program while implementing the comfort model to minimize discomfort. Geometry constraints are imposed on several points on the vehicle contour to ensure that the bus stays within the road bounds and docking is performed safely. The offline solution is then used as a reference in a real-time control event using a Volvo 7900 autonomous bus. To gain perspective, the measurements from the field test were compared to those of a human-driven trajectory. The results indicate that only around$\mathbf{\SI{1.7}{\%}}$of bus occupants will experience discomfort with the proposed model, in comparison to$\mathbf{\SI{60}{\%}}$in a human-driven bus, while respecting road and vehicle constraints and accurately docking within an acceptable predefined distance from the curb. We also show through simulations that in comparison to our proposed model, the traditional method of quadratic penalties for comfort modeling produces higher discomfort levels and prevents some trajectories characterized by high acceleration and jerk. Such trajectories can be permitted and comfortable using our approach.

Images from a front-facing camera on a vehicle can be used to estimate the varying Road Surface Conditions (RSC) ahead to warn the driver or to initiate automatic speed reduction in slippery road conditions. Previous works have successfully used deep-learning models to identify the RSC in the ego lane. Here, we focused on developing a model for predicting the RSC in multiple lanes simultaneously, relevant if changing lanes is an option. The proposed model estimate the RSC on the ego lane as well as in the adjacent lanes only if the adjacent lanes exists in the image. Furthermore, a data set is developed using more than 12,000 images from public benchmarks and privately captured images to facilitate multi-lane RSC estimation. Each image is assigned three RSC labels: with one for the ego, left and right lanes. The classes used are dry, wet, snow and snow-tracks. Our analysis with several network architectures has revealed that the model is capable of estimating the RSC in adjacent lanes with a similar level of performance as of the ego-lane.

This paper considers a coordination problem for Connected and Automated Vehicles (CAVs) in mixed traffic at unsignalized intersections. In such a setting, the behavior of the Human-Driven Vehicles (HDVs) is difficult to predict, thus challenging the formulation and the solution of the coordination problem. To solve this problem, we propose a coordination strategy, where CAVs are used as both sensors and actuators in mixed platoons. A timeslot-based approach is used to coordinate the occupancy of the intersection and to compensate for the HDVs behavior. The proposed approach has a bi-level optimization structure built upon the Model Predictive Control (MPC) framework that decides the crossing order and computes the vehicles’ commands. In simulations, we show that the choice of the HDV prediction model heavily affects the coordination by evaluating the performance of two different HDV models: car-following and constant velocity, where the latter demonstrates more consistent results in the presence of deviation of the HDVs’ behavior from a nominal model.

We propose an eco-driving technique for a heavy-duty ego vehicle that drives behind a leading heavy-duty vehicle. By observing a decrease in speed of the leading vehicle when driving uphill, its power capability is estimated and its future speed is predicted within a look-ahead horizon. The predicted speed is utilised in a model predictive controller (MPC) to plan the optimal speed of the ego vehicle such that its fuel consumption is minimised, while keeping a safe distance to the leading vehicle and reducing the need for braking. The effectiveness of the proposed technique is analysed in two case studies on real road topographies. By using the leading vehicle observer, fuel savings are achieved up to 8% compared to the case where the preceding vehicle is assumed to have a constant speed within the look-ahead horizon.

This paper studies the problem of optimal overtaking of a slow-moving leading vehicle in the presence of oncoming and/or adjacent vehicles with varying but known longitudinal speeds. A computationally efficient modeling approach is introduced, in which the overtaking problem is formulated by sampling in relative distance to the leading vehicle, replacing velocity state with its inverse and utilizing a nonlinear change of control variables. These three steps achieve a computationally efficient nonlinear control problem that can be solved using the sequential quadratic programming. Measures have been taken to ensure the feasibility of the nonlinear problem, even when the sequential quadratic programming iterates are stopped prematurely. A case study is presented, where this new formulation is compared with the previously published formulations in terms of solution quality and computation time.

This paper investigates the impact of the communication frequency on the remote control of automated vehicles. In particular, we consider a remote controller, which receives vehicles' state information and issues control commands based on a model predictive control (MPC) framework, to steer the vehicles to reach their respective target position intervals at given specific times. We present a framework where both state information (from the vehicles to the controller) and control actions (from the controller to the vehicles) are communicated through a wireless network. Due to limited communication resources and possible channel impairments, information is not necessarily always provided to the destination (either the controller or the vehicles). Herein, we particularly focus on the communications to the controller and investigate the effect of frequency and last instant of communication. Our results quantify the impact of these factors on the system performance, and subsequently, underline the need for an efficient resource allocation scheme.

This paper deals with the problem of human-robot cooperative object manipulation for cases where the grasp position of the operator can change during task execution similar to human-human collaborative scenarios. In state of the art algorithms for cooperative object handling, a constant grasping position is considered for the operator. In order to accommodate the changes of the human grasping point in the control design, we do not depend on sensors on the operator's hand or on the object but we employ estimates obtained through a recursive least-squares estimator. The estimation algorithm uses only the measured wrenches obtained by a force/torque sensor located at the end-effector of the manipulator. We also propose a switching strategy for a damping controller based on the online estimates. Simulation results are provided in order to demonstrate the proposed method.

We consider the problem of controlling a vehicle moving towards an intersection by means of a remote controller over an unreliable channel. This channel affects both uplink communication (when the vehicle sends its state information to the controller) and downlink information (when the vehicle receives control actions from the controller). We propose a probabilistic framework to compute control actions at the controller in the presence of such unreliable communications. The controller is evaluated under different channel conditions and compared to two nominal controllers, one that assumes perfect communication and one that assumes no communication. We find that for low packet loss rates, the proposed controller leads to less aggressive control actions than the former and generally lower cost than the latter. We additionally consider the mismatch between the perceived knowledge of the channel at the controller, and the actual channel conditions. We evaluate the performance of our controller under this mismatch, which is of interest when the controller is designed.

This paper investigates the hypothesis that threat assessment algorithms, developed for passenger cars and modified for articulated vehicles in low-speed scenarios, can be used to avoid a majority of the collision accidents in construction environments. The effect of using a collision avoidance system on a Volvo A25 hauler is investigated in several simulated scenarios with stationary obstacles. It is concluded that the safety benefit is severely limited by the relatively low braking capacity of the machine. Due to the limited capacity, braking manoeuvres must be initiated rather early, therefore steering is more efficient in many situations. Moreover, collisions are difficult to avoid when there is an offset between the machine and obstacle which is likely to occur in unstructured environments. The effect of different driving styles is also analysed and adapting the collision avoidance function to the individual driver can increase the number of collisions prevented. Another possible way to increase the safety benefit is to introduce a safety margin around obstacles. Both these options, however, increase the risk for generating false alarms.

This paper presents an algorithm for strategic decision making regarding when lane change and overtake manoeuvres are desirable and feasible. By considering the task of driving on two-lane, one-way roads, as the selection of desired lane and velocity profile, the algorithm provides useful results in terms of velocity control as well as a decision variable corresponding to whether a lane change manoeuvre should be performed. The decision process is modelled through a mixed logical dynamical system which is solved through model predictive control using mixed integer program formulation. The performance of the proposed control system is explored through simulations of varying driving scenaria on a two-lane, one-way road, which shows the capability of the system to achieve appropriate longitudinal and lateral control strategies depending on the traffic situation.

In this paper, we investigate predictive approaches to the problem of roadway departure prevention via automated steering and braking. We assume a sensing infrastructure detecting road geometry and consider a two-layer accident avoidance framework consisting of a threat assessment and an intervention layer. A novel active safety function for prevention of loss of vehicle control is proposed and implemented using the considered accident avoidance framework. Simulation and experimental results are presented, showing that the proposed approach effectively exploits road preview information to prevent the vehicle from operating in regions of the state space where standard electronic stability control systems are normally activated.

We propose two model-based threat assessment methods for semi-autonomous vehicles, i.e., human-driven vehicles with autonomous driving capabilities. Based on information about the surrounding environment, we introduce a set of constraints on the vehicle states, which are satisfied under “safe” driving conditions. Then, we formulate the threat assessment problem as a constraint satisfaction problem. Vehicle and driver mathematical models are used to predict future constraint violation, indicating the possibility of accident or loss of vehicle control, hence, the need to assist the driver. The two proposed methods differ in the models used to predict vehicle motion within the surrounding environment. We demonstrate the proposed methods in a roadway departure application and validate them through experimental data.

This paper presents a model-based algorithm that estimates how the driver of a vehicle can either steer, brake, or accelerate to avoid colliding with an arbitrary object. In this algorithm, the motion of the vehicle is described by a linear bicycle model, and the perimeter of the vehicle is represented by a rectangle. The estimated perimeter of the object is described by a polygon that is allowed to change size, shape, position, and orientation at sampled time instances. Potential evasive maneuvers are modeled, parameterized, and approximated such that an analytical expression can be derived to estimate the set of maneuvers that the driver can use to avoid a collision. This set of maneuvers is then assessed to determine if the driver needs immediate assistance to avoid or mitigate an accident. The proposed threat-assessment algorithm is evaluated using authentic data from both real traffic conditions and collision situations on a test track and by using simulations with a detailed vehicle model. The evaluations show that the algorithm outperforms conventional threat-assessment algorithms at rear-end collisions in terms of the timing of autonomous brake activation. This is crucial for increasing the performance of collision-avoidance systems and for decreasing the risk of unnecessary braking. Moreover, the algorithm is computationally efficient and can be used to assist the driver in avoiding or mitigating collisions with all types of road users in all kinds of traffic scenarios.

This correspondence concerns the estimation algorithm for hinging hyperplane (HH) models, a piecewise-linear model for approximating functions of several variables, suggested in Breiman (1993). The estimation algorithm is analyzed and it is shown that it is a special case of a Newton algorithm applied to a sum of squared error criterion. This insight is then used to suggest possible improvements of the algorithm so that convergence to a local minimum can be guaranteed. In addition, the way of updating the parameters in the HH model is discussed. In Breiman, a stepwise updating procedure is proposed where only a subset of the parameters are changed in each step. This connects closely to some previously suggested greedy algorithms and these greedy algorithms are discussed and compared to a simultaneous updating of all parameters.