Abstract Complex and changing driving environments not only affect the operating requirements of automatic wheel loader but also threaten its driving safety. Therefore, the automatic wheel loader must adopt appropriate braking strategies to realize accurate control of the brake pedal aperture under certain operating conditions. For the V-shaped operation mode of wheel loader, the operator's operation specification is evaluated using three characteristics: operation time, driving distance and friction work. By combining the driving data of experienced drivers in different driving environments with deep learning, a deep long short-term memory network was constructed to predict the brake pedal aperture for different braking types. The proposed anthropomorphic control method that combines driving data and deep learning can be used to predict the aperture value of the wheel loader brake pedal in complex driving environments. This would enable the braking process to conform to the braking decisions of experienced drivers and thereby meet the operational requirements while ensuring driving safety.

中文翻译：

---

基于深度学习的轮式装载机制动踏板孔径预测

摘要 复杂多变的驾驶环境不仅影响自动轮式装载机的运行要求，而且威胁其行驶安全。因此，自动轮式装载机必须采用适当的制动策略，才能在一定工况下实现对制动踏板开度的精确控制。对于轮式装载机的V型作业方式，从作业时间、行驶距离、摩擦功三个特征来评价操作者的作业规范。通过将经验丰富的驾驶员在不同驾驶环境下的驾驶数据与深度学习相结合，构建了一个深度长短期记忆网络来预测不同制动类型的制动踏板开度。所提出的结合驾驶数据和深度学习的拟人控制方法可用于预测复杂驾驶环境下轮式装载机制动踏板的孔径值。这将使制动过程符合经验丰富的驾驶员的制动决策，从而在确保驾驶安全的同时满足操作要求。