WirelessCar3 years ago
Autonomous Mobility as a Service (MaaS)
Autonomous Mobility as a Service (MaaS), and the connected car services that will determine your successShare
WirelessCar3 years ago
How does machine learning contribute to WirelessCar’s development of connected car services?
Machine learning plays an essential role in the development of connected cars and the digital services that allow these software-defined vehicles to reach their full potential. In this article, we take a closer look at WirelessCar’s work with machine learning, how that work contributes to new or improved connected car services, and the role of the OEMs in this process. What is machine learning? Machine learning is a type of artificial intelligence where computer algorithms are trained to learn from sets of data, instead of being hard coded (thus finding the right strategy from the training data). We use the datasets to train machine learning models. In “traditional programming”, a programmer will break down an issue or problem, find the underlying patterns, and instruct a program on how to solve the issue. With machine learning, we provide the model with all the relevant input, as well as the correct results we want it to predict. The model will then work out the pattern, what is causing the issue/problem, find a way for accurately predicting the results and use that to generate answers. How can machine learning improve connected car services? In the field of connected cars and connected car services, machine learning is particularly useful for data analysis. Connected cars and their sensors provide enormous amounts of data; data which can be analyzed and used to improve virtually every aspect of car ownership, fleet management, and mobility at large. Some examples: • Anomaly detection – machine learning can help identify issues by detecting unusual behavior. By training on both normal and failure patterns using DTC (Diagnostic Trouble Code) reports, it is able to identify deviations, which can result in quicker diagnoses, improved vehicle reliability, and reduced maintenance costs. • Better digital services for drivers – drivers’ preferences and user patterns (while not collected or analyzed in individual detail) can contribute to making connected car services better and more accessible; through optimized routing or improved EV charging, for example. • Safety and security – machine learning helps detect risks to driver safety and automotive cybersecurity. Maintenance needs can also be predicted and met more efficiently. WirelessCar’s data science and machine learning processes So, what does the greater data science process look like in our case, and where does machine learning come into the picture? It all starts with the identification of a problem – an issue that needs to be solved. Depending on what the problem is, and the connected car data at our disposal, we determine which machine learning technique we should use (more on that below). Once collected, the connected car data is prepared, analyzed, and processed to fit the machine learning technique we want to use. The data and the machine learning model must be relevant to the task at hand, so that they can provide actual answers to the problems we seek to solve. The machine learning models are then put to work, a process known as training. The models receive the prepared data, find patterns, make predictions, and learn from it – getting better as they go along. Once we have a result, we analyze it and discuss with the stakeholders involved on how best to apply it in our work. Which machine learning techniques does WirelessCar use? WirelessCar currently employs all the three main types of machine learning: supervised learning, unsupervised learning, and reinforcement learning. Supervised learning is used when we are trying to train a model to solve an issue using labeled data (thus data tagged with known, correct information). This is a good method to use as long as the data is of good quality and relevant to the problem we are trying to solve. A concrete example: Engine failure detection, based on diagnostic trouble codes (DTC) with known failure and reparation dates. Unsupervised learning, by contrast, is used when we need to come to a conclusion by working with unlabeled data. The machine learning model is required to analyze and find patterns from this data, such as when we try to categorize drivers into different groups according to certain parameters. We may not initially know how many groups there will be, nor how best to divide them up. Through unsupervised learning, we can find and evaluate the type of clustering/categorization that makes the most sense for our specific use-case. Then there is reinforcement learning, which means that the machine learning model learns through continuous feedback on what it does. Take autonomous driving: the model can be given rewards or penalties depending on how well it performs/”drives”. If it drives into an obstacle, it will get minus points and thus learn to avoid doing that in the future, for example. Ensuring the security and reliability of our machine learning work WirelessCar has chosen to work with Amazon Web Service (AWS) as our data science and machine learning partner. Our partnership with AWS goes back a long time, and our data is stored in AWS’s secure cloud services. We use Amazon SageMaker and Redshift for our machine learning models and QuickSight for visualization. It is crucial to stress that the data we analyze is anonymized. We do not have any traceable user information (drivers’ social security numbers, vehicle numbers, etcetera) and thus cannot use it in our machine learning. Additionally, our machine learning work does not involve that kind of granular breakdown of data. We are interested in the bigger picture, and in creating products and solutions that are useful for all OEMs, in all their markets. Overcoming the challenges of machine learning, together with OEMs What are the main challenges of machine learning, at least in this context? How can we overcome these challenges, and deliver better connected car services for OEMs? 1. Connected car data collection and analysis Essentially, data quantity is the key to connected car service quality. The more connected car data you can work with, the better analyses you will be able to make through your machine learning processes. Not all data will be relevant for a particular machine learning project, but as connected car services inevitably become more interconnected and complex, it is critical to have access to as much data as possible when developing these services. 2. Handling big data volumes Even comparatively small data sets can concern millions of vehicles, so handling enormous amounts of data is highly complex. At WirelessCar, we make sure to only invest in and work with the kind of technology that can live up to these demands, and that provides us with the tools necessary for big data management. That is a prerequisite for continuously enhancing our machine learning, and developing our connected car services for (and alongside) OEMs. 3. Collaborating closely with OEMs The closer the collaboration between us and the OEM, the better the results. Rarely is this more true than in the case of machine learning. We need to understand our customers’ needs and problems (and their customers’, in turn), and train our machine learning models accordingly. Through this collaborative process, not only are we able to overcome multifaceted issues together, but better prepare for the future as well.Share
WirelessCar3 years ago
How WirelessCar works with PI planning – lessons learned, and how we wish to inspire others
What is PI planning, and how does it benefit your organization? The main purposes of PI planning (Program Increment Planning) are to ensure the alignment of agile teams, create a shared vision, and to deepen the relationships between the teams and team members. In practice, that means that all the teams – across the organization – get together in a large room or auditorium, where they discuss (and visualize) their planning and how to deliver the most value to their customers. So, what are the most comprehensive benefits of PI planning? How do they strengthen the teams and the company organization? Build trust in and among your agile teams Working towards common goals requires trust: among teams and coworkers, between the staff and the leadership. The greater the number of interdisciplinary teams, the greater the need to build and reinforce that trust. By meeting regularly face-to-face, different teams grow to understand one another and learn about their respective processes. Your agile teams are able to plan their work better and with greater confidence Another strength of PI planning is that the teams lay out their work plans themselves. Along with a product owner (which helps them prioritize), each team decides on the content of their plan and how best to execute it. That in turn creates a sense of commitment to care for what you have planned, and make it happen. At the end of the PI planning session, each team has a vote of confidence about their plan. As we have continued to work with PI planning, we have seen that this confidence keeps growing among the teams. When you know where you are going, and why, that helps build trust, dedication, and ambition. Facilitate and improve decision-making Connected cars and connected car services are only getting increasingly complex. PI planning allows both OEMs and connected car service providers to make more well-informed decisions more quickly, and to prioritize correctly. Agile teams can execute faster When their development processes can be streamlined, agile teams can plan better, execute faster, and find better solutions to existing and potential issues. As the teams improve and learn through frequent evaluation and planning, they start providing even more customer value and an even greater user experience. How does WirelessCar work with PI planning? WirelessCar began working with the SAFe framework in 2017. PI planning is a key element of SAFe, and we found it to be an attractive, relevant model for us to work with. This especially as we wanted to work in a truly agile, scalable way. PI planning events require a lot of preparation, at least initially, in order to yield useful results. We certainly went through an initial learning curve. Today, however, PI planning is an established, appreciated aspect of how we work. These events are held at the WirelessCar headquarters in Gothenburg, Sweden, approximately every twelve weeks. We have gone from big room planning, face-to-face, to 100% remote and today a hybrid version. So while not everyone comes to our HQ, more and more teams see the value and importance of coming together for our PI planning sessions. We have adapted over time, and found a model that works very well for the entire organization. At the time of writing, we recently arranged our 25th PI planning event. The main objective is to work out how to deliver greater value to our customers, and how to do so more efficiently. This goes beyond setting new goals; in addition to that, we need to work out internal dependencies and synchronize development cycles between our teams. Issues get worked out, potential risks are raised and debated, priorities determined, and procedures can be changed to better serve ours and our customers’ needs. Our management team always participates, and its members meet with the various agile teams between PI planning events to ensure a good dialog. Inspiring customers and partners in their PI planning processes As our PI planning events became an integral part of our work at WirelessCar, our customers and partners began to take notice. Some of them had their first PI planning events together with us, in Gothenburg, before implementing this way of working in their own organizations. They recognized the efficiency and transparent nature of PI planning, and many have successfully applied it to their own operations. We have, in turn, been successful in better synchronizing our work with theirs, so that our organization is even more in tune with those of our customers and partners. We are able to deliver more products and solutions quicker than before, and respond better to the changing requirements of the automotive industry. What we have learned from our PI planning work, and what it means for us and the OEMs Your success with SAFe and PI planning will largely depend on how you deal with the issues you will encounter as part of the process. We are happy to say that we have learned and improved from ours at WirelessCar. Initially, PI planning events had a lot more stress and pressure to them. Today, they are more relaxed, and widely appreciated among all of our teams. Our investments in autonomous teams have lowered the number of dependencies, and facilitated the dialogue between each team and the management team. Also, there are less surprises during the PI planning events themselves, as we work even better together, within and among the teams. Some examples of areas where we have made progress: • Dependency-related complications between teams were more difficult to solve before we began working with PI planning. Today, these issues (to the extent they still exist) are often worked out before each PI planning events. •In the beginning, many teams brought up possible risks and escalations to the management team. Like the complications outlined just above, however, these issues increasingly get solved by the teams themselves nowadays: before or during our PI planning sessions. • It is difficult to overestimate the value of teams continually meeting each other and different stakeholders. PI planning events should not just be valuable, but enjoyable – a reason to celebrate. When we were not able to meet physically, we lost some of that, but we have now returned to a format that is more appreciated, and more in line with who we are as a company. These are some of the other key lessons we have learned from our work with PI planning: • Faster, better development and delivery of connected car services and digital products. • Closer cooperation with OEMs that have also chosen to work with PI planning. • Closer cooperation between interdisciplinary teams, and an even stronger relationship between them and the leadership team. • An active leadership that encourages ongoing conversation, and acts to make improvements where needed, will make your PI planning successful and rewarding. • PI planning contributes to a culture of transparency that benefits OEMs as well as our internal organization. • Greater focus on essential matters, rather than minor details. • A clearer vision of where we are headed and why, and what that requires of our teams and of us as a company. The trend we see now is that more and more teams work out complete and thorough plans after one day of PI planning. We also see the benefit of sometimes revisiting plans in a mid-PI review, in order to better adapt to the needs of our customers and our organization.Share
WirelessCar5 years ago
What is it like to work as a developer at WirelessCar?
Connected cars have been around for many years. However, we have only just begun to see this technology being implemented on a larger scale. As connected cars become more common, better and more differentiated services will be needed in order to meet customers’ demands. At WirelessCar, we get to create and develop these services for the world’s car manufacturers. As one of the company’s architects and developers, I would like to tell you more about what working as a system developer or application engineer at WirelessCar is like. What programming languages can developers expect to work with at WirelessCar? While Java certainly gets a lot of backend work done, we tend to adapt and alternate, depending on the configuration. Personally, I like to mix it up with Python in some cases. We use Node.js in certain contexts, and have at least one team working primarily in C#. On the frontend side, TypeScript is often used for new development, but we work with JavaScript as well. What about the developer’s freedom to choose which programming languages he or she gets to work with? Here I would say that the most important thing is that you are a good team player. If I were, say, a Kotlin expert, I would still have to be mindful of colleagues who do not have that same expertise; thus not drop a bunch of Kotlin work in their laps before going on vacation, for example. If you believe that a certain language will be beneficial to us, why not start out by introducing it in a small-scale solution. A test bot, perhaps, or a non-production system. If it does not work, no damage has been done, plus you probably learned something in the process. Now, if it turns out to be useful, we can develop and enhance your solution once it has proven to be powerful, reliable, and easy to maintain. Building connected car services for a wide variety of users At WirelessCar, we build a large chunk of the digital infrastructure between the car and the user. These backend cloud services, that the car communicates with, make up our core business. We also do frontend and mobile app work, and make sure that these solutions speak with our backend cloud services in a smooth and stable way. Individual drivers or car owners are not the only users we need to think about when we develop our services. Car dealers, connected fleet managers and call center operators are but some of the many user groups that make up the cycle of connected car services. Developing services that work end-to-end requires coordination; between us and the car makers, us and the partners we integrate with, and between our backend and frontend teams. One challenge you will encounter as a developer at WirelessCar is that of scale. Developing and implementing services that work for a few cars, or even a million, is one thing. As the scale grows to include tens or hundreds of millions of cars, there will be more factors to take into account and a great deal of complexity to wrap your head around. For example, a car sending messages to us a few times a day is nothing. Hundreds of millions of cars, on the other hand, means tens of thousands of messages per second, at peak time. Using reliable cloud services that work over time WirelessCar uses cloud services to provide efficient solutions for our customers. As these solutions need to be available and useful for years, we must work with suitable and reliable tools, and avoid technological fads. We currently do most of our work in Amazon Web Services (AWS). However, we use Microsoft Azure more and more, and expect the list of cloud services we use to keep growing. Starting your job as a WirelessCar developer While certain routines and work methods differ between our teams, our open and agile work process defines who we are as a company, and how we work. Most of our teams tend to work in two-week sprints. After every six or seven sprints, there will be cross-company coordination to check that the various teams do not have conflicting plans or priorities. When you start working at WirelessCar, you choose which operating system you want to use on your laptop: Linux, macOS or Windows. Some may see this as a minor detail, but personally, I am glad that we get to choose for ourselves. It is essential that you comply with our security practices; other than that, there are not many strict rules to abide by. I think that the workplace atmosphere at WirelessCar is characterized by activity, curiosity, and an eagerness to improve and grow. Ours is an industry where there is a lot going on all the time, so it is important to be alert and think ahead. Lastly, a confession: I am more interested in development than cars. Even so, cars happen to provide very good reasons for developing new services that solve problems. Solving those problems in an elegant and efficient way can be very satisfying, a bit like completing a complex puzzle. Fortunately, I get to work on these connected car services and problem-solving solutions every day. Written by Jens Andersson.Share
WirelessCar5 years ago
Smart EV Routing – A dynamic digital product for optimized EV travel
EV route planning enables drivers of electric vehicles to optimize their long-distance journeys. Continuous updates, based on real-time telemetry data, provide the driver with accurate, relevant information whenever he or she needs it. This results in a smoother ride for the driver. It also addresses and counteracts many common customer concerns related to electric vehicles. It is a golden opportunity for car makers to strengthen the relationship with their EV customers, and facilitate the shift to electric vehicles. What is an EV route planner? An EV route planner uses a variety of parameters in order to plan the best possible route for an EV driver. This is particularly useful for long-distance travel, as many drivers still express concerns about the availability of EV charging stations, range anxiety, the time needed for charging, etcetera. Thanks to the EV route planner, the driver gets to enjoy the journey and the EV without having to worry about these issues. The greater the intelligence of the EV route planner, the greater the value for the driver. The more data that can be collected – from both inside and outside of the car – the smarter and more accurate the information going back to the driver will be. How to optimize journeys for drivers of electric vehicles There are numerous factors that affect an EV journey, such as vehicle systems status, battery status, topography, weather, and road conditions like traffic and construction work. By taking all of these factors into account, the EV route planner is able to optimize the journey in a comprehensive way. That includes informing the driver about where to charge the vehicle, and for how long. Detours and worries about finding a charging station in time can thus easily be avoided. It is crucial to distinguish EV route planners from traditional navigation systems. Whereas the latter merely directs drivers to their destinations, an EV route planner is much more dynamic. The continuous collection of dynamic car data allows it to gather and present information about the EV and its batteries to the driver at all times. Moreover, it can optimize the charging process before and throughout the trip, as outlined above. It is also able to interact with dynamic charging infrastructure, to find out whether a charging station is in fact available and working, for example. How can car makers benefit from EV route planners? EV route planning supports car makers in showcasing and positioning themselves as credible EV providers. It is a concrete way of showing customers that the company is working actively to make EV travel easier and more enjoyable. Addressing customers’ concerns will also facilitate EV adoption, and likely increase customer loyalty. Consequently, it can strengthen the brand as well as its position within the EV segment. Depending on the strategy and business model, an EV route planner can be used for direct or indirect revenue streams, such as through monetized digital services. Smart EV Routing combines automotive industry experience with flexibility WirelessCar has more than twenty years of experience developing, delivering and operating secure digital solutions and products for connected cars. Smart EV Routing is a product that has been created to suit any car maker eager to provide their EV customers with a state-of-the-art route planner. Smooth, reliable EV route planning improves drivers’ journeys and helps them overcome common concerns related to EV long-distance travel. Smart EV Routing is an agnostic route planning product in the sense that it allows car makers to use whatever content providers they want for their specific solutions (regarding map data, traffic data, charging infrastructure, weather data and forecasts, and so on). It is important to remember that limiting an EV route planner by excluding certain charging infrastructure providers, for example, will not help drivers. They will want a comprehensive product or solution that caters to their needs, so car makers need to make sure that that is what they are offering their customers. WirelessCar packages and sends the vehicle data to our cloud service, and provides an optimized route plan in return. The route plan is provided to our OEM customer through an API for use and integration on- and/or off-board. Through data analytics, we are also able to help both car makers and charging infrastructure providers understand EV drivers’ patterns and decisions during longer trips. The potential of this technology is vast, and we should expect EV route planners to become increasingly capable and intelligent in the future. It is an exciting journey to be on, and a topic that we will doubtlessly return to in upcoming articles. Written by Rasmus Cornér.Share