مقایسه الگوریتم بهینهسازی چند منظوره کشف راهبردهای رانندگی با انسانها
|کد مقاله||سال انتشار||تعداد صفحات مقاله انگلیسی||ترجمه فارسی|
|5842||2013||9 صفحه PDF||28 صفحه WORD|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Expert Systems with Applications, Volume 40, Issue 7, 1 June 2013, Pages 2687–2695
- کارهای مرتبط
- شبیهساز رانندگی وسیلهی نقلیه
- کشف راهبردهای رانندگی با یک الگوریتم بهینهسازی
- بازنمایی راهبردهای رانندگی MODS
- الگوریتم بهینهسازی چند منظوره برای کشف راهبردهای رانندگی
- واسط کاربر
- آزمایشها و نتایج
- جدول 1: بهترین گسستهسازی فضای حالت، کنترلهای رانندگی، وزنهای مصرف سوخت، و تعداد گامهای پیشبینی
- جدول 2: مقادیر پارامتر الگوریتم تکاملی سطح بالاتر MODS
- جدول 3: مقادیر عینی راهبردهای رانندگی نشانهگذاری شده در شکل 7
When a person drives a vehicle along a route, he/she optimizes two objectives, the traveling time and the fuel consumption. Therefore, the task of driving can be viewed as a multiobjective optimization problem and solved with appropriate optimization algorithms. The comparison between the driving strategies obtained by humans and those obtained by the algorithms is interesting from several points of view. For example, it is interesting to see which strategies are better. To perform the human versus machine test, we compared the driving strategies obtained by the multiobjective optimization algorithm for discovering driving strategies (MODS) with those obtained by a group of volunteers operating a vehicle simulator. The test was performed using data from three real-world routes. The results show that MODS always finds better driving strategies than the volunteers, especially when the fuel consumption is to be reduced. Moreover, the results show that some volunteers always drive similarly in terms of traveling time and fuel consumption while others significantly vary their driving strategies.
When a person drives a vehicle along a route, he/she usually optimizes two objectives: the traveling time and the fuel consumption. Vehicle driving by minimizing only the traveling time is quite intuitive and straightforward: the vehicle has to be driven at the maximum allowed velocity all the time. On the other hand, when the fuel consumption has to be reduced, people usually follow some well-known guidelines (Johnson, 2006 and Weinger, 2007). However, even if a person follows these guidelines, the optimal driving strategies may not be obtained. To discover how good the human driving strategies are, an optimization algorithm can be designed and evaluated by comparing the obtained driving strategies with the human driving strategies. An example of an optimization algorithm for this problem is the multiobjective optimization algorithm for discovering driving strategies (MODS) (Dovgan et al., 2011 and Dovgan et al., 2011) that we designed and implemented. The algorithm was tested on data from real-world routes and the obtained driving strategies are better than the driving strategies found with previously used optimization algorithms (Dovgan, Tušar, Javorski, & Filipič, 2012), i.e., predictive control (Del Re, Allgower, Glielmo, Guardiola, & Kolmanovsky, 2010) and dynamic programming (Hellstrom et al., 2010 and Hellstrom et al., 2009). However, the driving strategies were not compared to the driving strategies pursued by humans. In this paper we compare the driving strategies obtained by MODS with the human driving strategies. The driving strategies were obtained by simulating the driving on data from real-world routes and minimizing both the traveling time and fuel consumption. To obtain the human driving strategies, an intuitive user interface was implemented and used by a group of volunteers. In addition, the volunteers were classified into categories as expert game players, regular computer users and occasional computer users. Afterwards, MODS driving strategies were compared to the human driving strategies on the individual basis and by taking into account specific categories. The paper is further organized as follows. Section 2 describes the related work in this field. Section 3 presents the vehicle driving simulator used to evaluate the driving strategies. The MODS algorithm is described in Section 4. Section 5 describes the user interface. Section 6 presents the experiments and the obtained results. Finally, Section 7 concludes the paper with the summary of work and ideas for future research.
نتیجه گیری انگلیسی
This study compared the driving strategies found by the multiobjective optimization algorithm for discovering driving strategies (MODS) with those found by the volunteers. The experiment used a vehicle driving simulator on data from three real-world routes, where MODS and 11 volunteers tried to find driving strategies with minimal traveling time and fuel consumption. The results show that on these routes MODS always find better driving strategies than the volunteers. Moreover, some volunteers always drive similarly in terms of the traveling time and the fuel consumption while others significantly vary their driving strategies. The volunteers were self-classified as expert game players, regular computer users or occasional computer users. The results indicate that two categories would be more appropriate for classifying the volunteers since the objective values of the driving strategies in the first two categories overlap. This indicates that subjective classification does not necessarily correspond with the quality of the produced driving strategies. This is in agreement with the findings of Canale and Malan (2002) who also found that the subjective classification of human driving styles is probably limited. Nevertheless, expert race game players found the best driving strategies among the volunteers, while regular computer users in the majority of cases found similar driving strategies as expert game players in terms of the traveling time but not the fuel consumption. The occasional computer users produced the worst driving strategies with significantly higher fuel consumption. In addition, the analysis of vehicle behavior for selected driving strategies revealed that low traveling time is obtained with a constant vehicle velocity which is the highest allowed, that volatile vehicle velocity increases the fuel consumption, and that the lowest fuel consumption is obtained with the pulse-and-glide driving strategy. In our future work we will compare human and MODS driving strategies on additional, more complex routes. Since the results show that the constant velocity is in most cases a good strategy, it would be interesting to add speed control to the vehicle simulator to simplify the human driving and obtain better human driving strategies. We could also postprocess the best driving strategies found by MODS, present them to humans in a compact and comprehensible form, and test whether they can then improve their own driving strategies.