Long-Term Vehicle Speed Prediction via Historical Traffic Data Analysis for Improved Energy Efficiency of Connected Electric Vehicles

Abstract

Connected and automated vehicles (CAVs) are expected to provide enhanced safety, mobility, and energy efficiency. While abundant evidence has been accumulated showing substantial energy saving potentials of CAVs through eco-driving, traffic condition prediction has remained to be the main challenge in capitalizing the gains. The coupled power and thermal subsystems of CAVs necessitate the use of different speed preview windows for effective and integrated power and thermal management. Real-time vehicle-to-infrastructure (V2I) communications can provide an accurate speed prediction over a short prediction horizon (e.g., 30 s to 60 s), but not for a long range (e.g., over 180 s). Therefore, advanced approaches are required to develop detailed speed prediction for robust optimization-based energy management of CAVs. This paper presents an integrated speed prediction framework based on historical traffic data classification and real-time V2I communications for efficient energy management of electrified CAVs. The proposed framework provides multi-range speed predictions with different fidelity over short and long horizons. The proposed multi-range speed prediction is integrated with an economic model predictive control (MPC) strategy for the battery thermal management (BTM) of connected and automated electric vehicles (EVs). The simulation results over real-world urban driving cycles confirm the enhanced prediction performance of the proposed data classification strategy over a long prediction horizon. Despite the uncertainty in long-range CAVs’ speed predictions, the vehicle-level simulation results show that 14% and 19% energy savings can be accumulated sequentially through eco-driving and BTM optimization (eco-cooling), respectively, when compared with normal driving (i.e., human driver) and conventional BTM strategy.

A comprehensive study of speed prediction in transportation system: From vehicle to traffic

“…The above vehicle information constrains the change of vehicle speed indirectly, and the dynamic vehicle states directly affect speed. These factors describe the running state of vehicle, including vehicle speed, acceleration, available fuel or power ( Sun et al., 2015a ), battery temperature ( Amini et al., 2020 ), transmission state ( Li et al., 2018a ), etc.…”

Integrated Power and Thermal Management of Connected HEVs via Multi-Horizon MPC

“…• an approximate forecast of the vehicle speed beyond the short-range accurate prediction window is also available. Such a forecast can be generated by processing traffic data collected from the connected vehicles traveling along the same route as the ego-vehicle, see [22], [23], [29].…”

“…The second segment is a shrinking horizon with low resolution to reduce the computation demands of the predictive controller. An approximate long-term prediction of the future vehicle speed, which can be realized by processing the traffic data collected from the connected vehicles [23], is incorporated in the second segment of the horizon.…”

Experimental Validation of Eco-Driving and Eco-Heating Strategies for Connected and Automated HEVs

“…As shown in previous studies [11][12][13][14][15], integrated power and thermal management (iPTM) of CAVs can greatly benefit from leveraging the coupling between power and thermal loads and accounting for the timescale separation between power and thermal dynamic responses. Along these lines, energy-efficient strategies for cooling (i.e., eco-cooling) of cabin [11,[16][17][18] and battery [13,[19][20][21], as well as iPTM strategies for co-optimization of engine, cabin, and aftertreatment systems [12,14,15,22,23] have been developed. When conflated with technologies focused on traction power optimization (e.g., eco-driving), efficient thermal management of CAVs is shown to have the potential for delivering fuel-savings of up to 18-20% [12,22,24].…”

“…This generated trajectory is (i) used to conduct eco-driving experiments, and (ii) leveraged as a look-ahead preview by MPC-based eco-heating strategy. Note that, other approaches have also been developed for short-and longterm predictions of vehicle speed using V2X [20,[27][28][29][30][31][32][33][34] which can be integrated with our eco-heating strategy.…”