Command modification using input shaping for automated highway systems with heavy trucks

Bae, Hong; Gerdes, J. Christian

doi:10.1109/acc.2003.1238913

Cited by 14 publications

(9 citation statements)

References 34 publications

(47 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Because of the increasing number of sensors in road vehicles, interest arose in the 1990s to identify vehicle parameters online [17], [21], [22]. Most of these earlier studies, both in simulation and experimentally, use an ICE vehicle as a case study [23], [24], [25], [26]. In the presented estimator algorithms, engine torque is often required to be measured.…”

Section: A Literature Overview: Online Rolling Resistance and Road Gr...mentioning

confidence: 99%

“…Except for engine or motor torque, additional sensors are often employed to base the estimation on. These include GPS [17], [18], [24], [27], which is often used to assess the road grade, or IMUs [19], [23], [26]. In most studies, the vehicle speed signal is also required and measured using wheel-speed sensors or a tachograph.…”

Section: A Literature Overview: Online Rolling Resistance and Road Gr...mentioning

confidence: 99%

See 1 more Smart Citation

Combined Rolling Resistance and Road Grade Estimation Based on EV Powertrain Data

Beckers

Besselink

Nijmeijer

2023

IEEE Trans. Veh. Technol.

View full text Add to dashboard Cite

Energy consumption prediction is increasingly important for eco-driving, energy management, and charging scheduling of electric vehicles. Detailed knowledge of the rolling resistance and road grade, combined here in a road-resistance profile, improves the accuracy of these predictions. This paper presents a recursive method to identify the position-dependent road-resistance coefficient using GPS position, powertrain power, and vehicle speed. The calculations make explicit assumptions regarding the spatial continuity of both road gradient and rolling resistance by defining road segments. A recursive least-squares method with Gaussian basis functions allows the estimates to be updated whenever a route segment is traversed anew. The method is tested on data gathered by a 12 m battery electric bus. The resulting road-resistance profile shows a strong resemblance to the road slope and captures changes in rolling resistance well, including a dependency on ambient temperature, which is in accordance with literature on tire rolling resistance. Including the resistance profile in a vehicle model reduces the error of the predicted powertrain power by 1.7 percent point compared to a conventional method, without the limitation of requiring a high-resolution digital elevation model.

show abstract

Section: A Literature Overview: Online Rolling Resistance and Road Gr...mentioning

confidence: 99%

Section: A Literature Overview: Online Rolling Resistance and Road Gr...mentioning

confidence: 99%

Combined Rolling Resistance and Road Grade Estimation Based on EV Powertrain Data

Beckers

Besselink

Nijmeijer

2023

IEEE Trans. Veh. Technol.

View full text Add to dashboard Cite

show abstract

“…Heavy‐duty truck are highly complex and their driving performance is details sensitive (Bae & Gerdes, 2003; Druzhinina et al, 2002; Kirches et al, 2013; Lattemann et al, 2004; Lu & Hedrick, 2005; Vahidi et al, 2003). Deep learning has been shown to capture details including those that are difficult to model or measure using state‐of‐the‐art physics‐based approaches (Bansal et al, 2016; Da Lio et al, 2019; James et al, 2020; Spielberg et al, 2019).…”

Section: Modeling Problem Formulationmentioning

confidence: 99%

Longitudinal deep truck: Deep longitudinal model with application to sim2real deep reinforcement learning for heavy‐duty truck control in the field

Albeaik

Vurimi

et al. 2022

Journal of Field Robotics

View full text Add to dashboard Cite

To develop cooperative adaptive cruise control (CACC), the choice of control approach often influences and can limit the choice of model structure, and vice versa. For heavy-duty trucks, practical application of CACC in the field is heavily influenced by the accuracy of the used model. Deep learning and deep reinforcement learning (deep-RL) have recently been used to demonstrate improved modeling and control performance for vehicles such as cars and quadrotors compared to state-of-the-art. The literature on the application of deep learning and deep-RL for heavy-duty trucks in the field, which are significantly more complex than cars, is still sparse, however. In this article, we develop a two-layer gray-box deep learning model to capture longitudinal dynamics of heavy-duty trucks while abstracting their complexity and present an approach to properly break the nested feedback loops in the model for training. We compare this model with three other alternative models and show that it achieves x ~10 better general performance compared to a standard artificial neural network and results in x ~4 and x ~40 slower steady-state acceleration and speed error growth rates, respectively. We then present an architecture to utilize these deep learning models within the deep-RL framework and use it to develop baseline CACC controllers that can be zero-shot transferred to the field. To carry out the work, we present a setup of differently configured trucks along with their interface architecture and stochastic driving cycle generators for data collection. Numerical validation of the approach demonstrated stationary and bounded modeling error, and demonstrated transfer of CACC controllers with consistent overshoot bounds and a stable approximately-zero steady-state error. Validation from field experiments demonstrated similarly consistent results. Compared to a state-of-the-art benchmark, the deep-RL controller achieved lower speed and time-gap error variance but higher time-gap error offset.

show abstract

“…Command shapers have been developed to filter real-time changes in desired velocity to reduce settling time to the new velocities and reduce the overall actuation needed to reach the desired state. One of the first applications of command shaping on automated convoys was focused on reducing actuator effort in heavy truck platoons, but the study did not focus on results with respect to velocity tracking and/or inter-vehicular spacing [10]. In this paper, the command shaper is designed to eliminate the "vibration" in the velocity and position of the follower vehicles.…”

Section: Remarkmentioning

confidence: 99%

Application of command shaping to a convoy of ground vehicles

Henderson

Feemster

Robertson

2011

2011 IEEE International Conference on Control Applications (CCA)

View full text Add to dashboard Cite

Abstract-This paper documents the effect that command shaping has on the inter-vehicular spacing performance of autonomous convoys. The control method is based only on distance measurements between vehicles in the platoon. The lead vehicle's longitudinal controller was developed to follow a given velocity. The follower vehicles maintain the desired spacing using distance measurements to the preceding and following vehicles. Three overall control schemes were experimentally tested: PD control without lead vehicle velocity information, PD control with lead vehicle velocity information and PID control without lead vehicle velocity information. In all cases adding command shaping led to improved performance. This approach leads to more physically realizable controller efforts for the follower vehicles and helps maintain and improve string stability.

show abstract

Command modification using input shaping for automated highway systems with heavy trucks

Cited by 14 publications

References 34 publications

Combined Rolling Resistance and Road Grade Estimation Based on EV Powertrain Data

Combined Rolling Resistance and Road Grade Estimation Based on EV Powertrain Data

Longitudinal deep truck: Deep longitudinal model with application to sim2real deep reinforcement learning for heavy‐duty truck control in the field

Application of command shaping to a convoy of ground vehicles

Contact Info

Product

Resources

About