This dissertation deals with traction control for electric vehicles. A new control method is proposed which makes full use of the inherent advantages of electrical traction systems to provide a novel general approach to the traction control of electric vehicles.

Due to growing concern about global environmental problems and shrinking non-renewable energy sources, research on electric vehicles and hybrid electric vehicles is gaining more and more attention. Meanwhile, significant improvements in power electronics, energy storage and control technology have made electric vehicles fully feasible, also paving the way for the introduction of innovative technologies that will give them an even greater advantage. In particular, the realization of practical in-wheel motors will allow for some revolutionary changes in the design of electric vehicles, with new topologies not only contributing to the energy efficiency and freedom of vehicle design, but also providing considerable benefits to the controllability of electric vehicles.

On the other hand, active safety control performs a very important role in modern vehicles. However, its core technology, traction control, is still problematic. Conventional traction control in internal combustion engine vehicles, which is dependent on the slip ratio reference, is limited in effectiveness and adaptability by the fact that chassis velocity and road conditions cannot be measured or detected in a practical, reliable way. For this reason, much research has been applied to controllers for electric vehicles, seeking to make use of their advantages to overcome the obstacles of calculating chassis velocity and detecting road conditions in real time to perform traction control. However, these control designs, which are based on compensation, compromise performance for system stability, and the compensation gain is tuned for some specific tire-road conditions, also limiting the practicability of these methods.

Therefore, this dissertation, making use of the advantages of electric vehicles, focuses on development of a core traction control system based on a proposed concept called Maximum Transmissible Torque Estimation, which requires neither chassis velocity nor information about tire-road conditions. In this system, use is made of only the torque reference and the wheel rotation to estimate the maximum transmissible torque to the road surface, then the estimated torque is directly applied to antislip control.

Chapter 1 introduces the research background, particularly the history and the advantages of electric vehicles, as well as the importance of traction control in vehicles. Chapter 2, using Anti-lock Braking Systems as an example, reviews conventional traction control and analyzes the difficulties it faces. Chapter 3 presents the novel topology of traction control and uses an equivalent model to provide the stability analysis of the control system. Chapter 4 describes the experimental electric vehicle used as a test-bed and presents the simulations and experiments performed on it to evaluate the proposed control system. Comparative experimental results and additional simulation results are then followed by a detailed discussion and analysis. Finally, Chapter 5 discusses the extension of the proposed traction control to a two-degree-of-freedom vehicle motion control system. Simulation results on an example system are followed by a detailed discussion of its effectiveness.

Based on thorough experimental tests and simulations, and supported by detailed theoretical analysis as well as wide acceptance in both domestic and international conferences and authoritative journals, the author firmly believes in the advantages of the proposed control topology. The controller not only has better antislip performance and higher adaptability in different tire-road conditions than previous controllers, but also has greater robustness to perturbations in vehicle mass and disturbances in driving resistance, which also demonstrates the high practicality of the method. From the viewpoint of implementation, only torque and wheel rotation are used as the input variables to the controller, which contributes not only to lower cost, but also higher reliability and greater independence from driving conditions.

These advantages qualify the proposed control method as a general approach for traction control, as well as a basis for more sophisticated and advanced motion control in electric vehicles. Moreover, the control philosophy implied in the proposed control topology merits further investigation and can be expected to find application as a general approach to a wider class of control problems.

本論文は,General Approach to Traction Control for Electric Vehicles Based on Maximum Transmissible Torque Estimation(最大伝達可能トルク推定にもとづく電気自動車の一般的なトラクション制御に関する研究)と題し,電気モータに本来備わっている高い制御性を十分に生かした電気自動車のトラクション制御として,一般的でかつ新しい制御手法の提案と構築を行った成果をまとめたもので,英文で記述された全6章から成る。

自動車のアクティブセイフティ技術の核となるトラクション制御には未だ多くの課題が残っているが,従来のスリップ率に基づく手法では,車体速度や路面状態の推定が必要で,その効果と適用性には限界がある。そこで本研究では,最大伝達可能トルクという概念を導入し,車体速度や路面状態情報を陽に利用しない推定法を提案,電気自動車の利点を最大限発揮できるトラクション制御システムを構築することとした。最大伝達可能トルクは,トルク指令値と車輪の回転速度から算出され,それを粘着制御に利用することで効果的なトラクションコントロールを実現するものである。

第1章では,本研究の背景,目的を述べている。とくに,電気自動車の発展の歴史や利点に触れ,電気自動車におけるトラクション制御の重要性を述べている。第2章ではABSを例に用いて,従来のトラクション制御を評価し,その問題点を分析している。第3章は,最大伝達可能トルクの推定にもとづく新しい粘着制御法を提案し,さらに等価モデルを用いて,システムの安定性解析を行っている。第4章では,実験用電気自動車の製作について述べ,実車実験とシミュレーションにより,提案した制御手法を評価している。様々な制御パラメータ,車体のパラメータを用いた多くの実車実験を行い,提案手法の有効性を確かめている。また,実車実験ができない場合においては,シミュレーションによる検討で補ったことを述べている。第5章では,提案したトラクション制御を二次元での車両運動制御へ拡張する構想について議論を行い,シミュレーションによってその効果を確認している。第6章は結論である。

提案したコントローラは,高いアンチスリップ制御能力と様々なタイヤ・路面状況への高い適用性,さらに車両重量の変動や走行抵抗などの外乱に対するロバスト性を有し,高い実用性を持っている。また,トルクと車輪の回転速度のみがコントローラへの入力として使われるので,低コスト化・高信頼化を実現できると考えられる。また,提案手法は,トラクション制御への適用にとどまらず,電気自動車のより高度で複雑な車体制御の基礎ともなり得るとしている。

以上これを要するに,電気自動車のトラクション制御において,最大伝達可能トルクという概念を導入し,車体速度や路面状態情報を陽に利用しない推定法をベースとする新しい粘着制御法を提案し,実際に電気自動車を製作してその有効性を実証したものであって,電気工学,自動車工学,制御工学上,貢献するところが少なくない。

よって本論文は博士(工学)の学位請求論文として合格と認められる。