Multi-Objective Comprehensive Control of Trajectory Tracking for Four-In-Wheel-Motor Drive Electric Vehicle With Differential Steering

Wang, Hongbo; Hu, Chenglei; Cui, Wei; Du, Haiping

doi:10.1109/access.2021.3074215

Cited by 6 publications

(3 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Nevertheless, the commonly-used global planners are unable to generate optimal paths because they search a feasible path from the map via a heuristic method. Moreover, the local planners need careful tuning for the coupled factors involving the cost weights of collision, path following, motion speed, etc.. For certain simple situations, if the environment does not change and the global path avoids all obstacles, some path tracking methods can be directly used to follow the path without the concern of collision [15], [16]. The aforementioned methods assume that a global map is already known, which is impractical when frequently changing the environment.…”

Section: Related Workmentioning

confidence: 99%

A Hierarchical Deep Reinforcement Learning Framework With High Efficiency and Generalization for Fast and Safe Navigation

Zhu

Hayashibe

2023

IEEE Trans. Ind. Electron.

View full text Add to dashboard Cite

We present a hierarchical deep reinforcement learning (DRL) framework with prominent sampling efficiency and sim-to-real transfer ability for fast and safe navigation: the low-level DRL policy enables the robot to move towards the target position and keep a safe distance to obstacles simultaneously; the high-level DRL policy is supplemented to further enhance the navigation safety. We select a waypoint located on the path from the robot to the ultimate goal as the sub-goal to reduce the state space and avoid sparse reward. Moreover, the path is generated based on either a local or a global map, which can significantly improve the sampling efficiency, safety, and generalization ability of the proposed DRL framework. Additionally, a target-directed representation for the action space can be derived based on the sub-goal to improve the motion efficiency and reduce the action space. In order to demonstrate the eminent sampling efficiency, motion performance, obstacle avoidance, and generalization ability of the proposed framework, we implement sufficient comparisons with the non-learning navigation methods and DRL-based baselines, with videos, data, code, and other supplemental material shown on our website 1 .

show abstract

Section: Related Workmentioning

confidence: 99%

A Hierarchical Deep Reinforcement Learning Framework With High Efficiency and Generalization for Fast and Safe Navigation

Zhu

Hayashibe

2023

IEEE Trans. Ind. Electron.

View full text Add to dashboard Cite

show abstract

“…To demonstrate the effectiveness of the proposed control strategy, simulations under the double lane condition (Wang et al, 2021) of different road adhesion coefficient ( μ = 0 . 7 , 1 . 2 ) and different vehicle speed ( v x = 30 , 45 , 60 km · h − 1 ) are conducted via CarSim-Simulink platform, and the main parameters of the vehicle are shown in Table 2. Figure 12 illustrates the path-following performance comparison with autonomous steering, yaw stability control (Rajamani, 2012) and coordination control at different speeds and road coefficients, which demonstrate that coordination control can obtain better tracking accuracy than autonomous steering and yaw stability control due to the cooperation of autonomous and differential steering.…”

Section: Coordination Control For Path-followingof Ddaevmentioning

confidence: 99%

Adaptive coordination control strategy for path-following of DDAEV with neural network based moving weight coefficients

Xie

Wang

et al. 2022

Transactions of the Institute of Measurement and Control

View full text Add to dashboard Cite

This paper presents an adaptive coordination control strategy for path-following of distributed drive autonomous electric vehicles (DDAEV). A model predictive control (MPC) algorithm is used to realize path-following through autonomous steering, where the prediction time is adaptive in relation to different driving conditions. Due to the dynamic characteristic of distributed drive vehicle, the differential torque control is also utilized based on the deviation of path to realize path-following. In order to make full use of the advantages of these two path-following methods, coordination control of autonomous steering and differential steering is adopted to improve the transient response speed, flexibility of steering system and path-following performance by setting moving weight coefficients based on neural network. Results of CarSim-Simulink co-simulation and real vehicle experiment both verify that the proposed coordination control can obtain steering flexibility, as well as tracking accuracy and reliability under various driving conditions.

show abstract

“…By changing the driving torques of the in-wheel motors, active braking control, electronic stability control and the adaptive cruise control can be carried out conveniently. Therefore, the EV-DFIM has good development in the future [3][4][5][6][7]. The EV-DFIM has precise and flexible torque response, superior performance in dynamic performance and handling activity, which makes it easier to adopt lane keeping technology.…”

Section: Introductionmentioning

confidence: 99%

Homogeneous Domination-Based Lateral Stability Control Method for Electric Vehicle Lane Keeping System

et al. 2022

View full text Add to dashboard Cite

Lane keeping control technology can improve driving safety. But the direct control torque, caused by four in-wheel motors of an electric vehicle driven by four in-wheel motors (EV-DFIM), will affect the vehicle lateral stability. For solving this issue, a two-degree-of-freedom (2DOF) lateral dynamic model of an EV-DFIM lane keeping system is constructed to obtain the lateral stability control state equations. Then a homogeneous domination-based state observer and output feedback controller are designed to improve the vehicle lateral stability and balance when the lane keeping system is working. The Lyapunov method is used to prove that the designed homogeneous domination-based output feedback controller can globally asymptotically stabilize the system. Then, a direct yaw moment strategy is presented based on the optimal allocation algorithm with the lowest utilization rate of the tire to allocate the desired torques for every inwheel motor. Finally, numerical simulation and HIL simulation are carried out to verify that the proposed lateral stability control method based on homogeneous domination theory for the EV-DFIM lane keeping system has a good control effect and robustness.

show abstract

Multi-Objective Comprehensive Control of Trajectory Tracking for Four-In-Wheel-Motor Drive Electric Vehicle With Differential Steering

Cited by 6 publications

References 43 publications

A Hierarchical Deep Reinforcement Learning Framework With High Efficiency and Generalization for Fast and Safe Navigation

A Hierarchical Deep Reinforcement Learning Framework With High Efficiency and Generalization for Fast and Safe Navigation

Adaptive coordination control strategy for path-following of DDAEV with neural network based moving weight coefficients

Homogeneous Domination-Based Lateral Stability Control Method for Electric Vehicle Lane Keeping System

Contact Info

Product

Resources

About