C AR MA: Towards Personalized Automotive Tuning ∗ Tobias Flach 1 , Nilesh Mishra 1 , Luis Pedrosa 1 , Christopher Riesz 2 , and Ramesh Govindan 1 1 University of Southern California 2 Rutgers University, Camden Abstract Wireless sensing and actuation have been explored in many contexts, but the automotive setting has received rel- atively little attention. Automobiles have tens of onboard sensors and expose several hundred engine parameters which can be tuned (a form of actuation). The optimal tuning for a vehicle can depend upon terrain, traffic, and road condi- tions, but the ability to tune a vehicle has only been avail- able to mechanics and enthusiasts. In this paper, we de- scribe the design and implementation of CARMA (Car Mo- bile Assistant), a system that provides high-level abstrac- tions for sensing automobile parameters and tuning them. Using these abstractions, developers can easily write smart- phone “apps” to achieve fuel efficiency, responsiveness, or safety goals. Users of CARMA can tune their vehicles at the granularity of individual trips, a capability we call person- alized tuning. We demonstrate through a variety of appli- cations written on top of CARMA that personalized tuning can result in over 10% gains in fuel efficiency. We achieve this through route-specific or driver-specific customizations. Furthermore, CARMA is capable of improving user satisfac- tion by increasing responsiveness when necessary, and pro- moting vehicular safety by appropriately limiting the range of performance available to novice or unsafe drivers. Categories and Subject Descriptors D.2.13 [Software Engineering]: Reusable Software— Domain engineering; J.7 [Computers in Other Systems]: Consumer products General Terms Design, Experimentation, Performance Keywords Automobile, Engine Control Unit, Scanning, Tuning ∗ This material is based upon work supported by the National Science Foundation under Grants No. CNS-021778, CCF-0830569 and CCF-1048606 and CCF-820230. Any opinions, findings and conclusions or recomendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the Na- tional Science Foundation (NSF). Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. SenSys’11, November 1–4, 2011, Seattle, WA, USA. Copyright 2011 ACM 978-1-4503-0718-5/11/11 ...$10.00 1 Introduction Networked sensing and actuation have been explored in many natural and man-made environments, most recently in smart-grid and smart-building technologies [42, 28]. The emergence of smartphones has added a new dimension to networked sensing, enabling context awareness [35] and par- ticipatory sensing [37]. However, one ubiquitous man-made artifact, the automobile, has received relatively less attention. Automobiles have tens of onboard sensors which can continuously monitor engine parameters such as the air flow, the current gear setting, the engine RPM and speed. More importantly, automobiles can be tuned by modifying engine parameters, such as RPM limits, transmission shift points, or air fuel mixtures. In general, automotive tuning is a com- plex and dangerous task, so it is performed today by trained mechanics, enthusiasts, or fleet operators. There are, however, benefits to being able to tune cars on a larger scale. Specifically, personalized tuning, in which a user tunes a car at a granularity of a trip, can potentially have significant fuel efficiency benefits. Most cars come with fac- tory default parameters and are rarely modified, but as with any parameterizable system, a single set of parameters is un- likely to be optimal across all settings. Indeed, the fuel effi- ciency of a car can vary widely with the route, the conges- tion along the route, terrain, and other factors (Section 2). Personalized tuning can determine route-specific engine pa- rameters to squeeze efficiency gains. Fuel efficiency is not the only motivation for personalized tuning. This capability can be used to improve safety or to increase the responsiveness 1 of the car when necessary. For example, it is possible to set speed and RPM limits on cars in order to prevent misuse or rash driving, a capability that is useful when handing off a car to a teenaged driver. This paper explores the design and implementation of CARMA, a system for smartphones that is designed to de- mocratize personalized tuning (Section 3). Because it runs on smartphones, CARMA is highly portable and makes trip-granularity tuning convenient. CARMA is also pro- grammable, enabling the development of smartphone apps that permit personalized tuning for flexibly achieving fuel 1 Responsiveness is the measure of how quickly and forcefully a vehicle reacts to changes in throttle position. This includes en- gine braking while off the throttle, and maintaining a lower gear for quicker acceleration during throttle increases.