 My name is Reem Malik, and in this presentation, I will discuss considerations in autonomous applications and how you can accelerate your development using Asina's inertial navigation solutions. Our world, today and especially tomorrow, is comprised of machines that move. Many of these machines need to move autonomously without a human to guide them. Underlying any autonomous technology is the promise and necessity of superior safety. The technology in an autonomous vehicle helps it navigate the world around it. To do it safely and precisely is of utmost importance. The increasingly sophisticated functions of AVs necessitate that each vehicle have reliable knowledge of its position. Perception sensors see the world around the vehicle and are often the primary source of information for active decision making and vehicle control. Perception systems include lidar, radar, cameras, infrared and others. These are all combined with massive compute power and deep learning algorithms. Guidance and inertial navigation systems inform vehicles where in the world they are and how to navigate to their destination. These systems consist of GNSS, GPS receivers and an INS inertial navigation system, which includes inertial motion sensors and inputs from odometry and steering sensors. An inertial measurement unit, or IMU, is an electronic module that integrates multiple inertial sensors to generate acceleration and angular rate measurements along multiple axes or degrees of freedom. A six degree of freedom IMU consists of three-axis gyroscope and a three-axis accelerometer. Measurements from these sensors taken over time are combined using an extended Kalman filter to make highly accurate calculations of position, velocity and attitude or orientation. Attitude heading and reference systems combine magnetometer readings with IMU data to calculate heading, roll and pitch. An INS adds GPS to track the position, orientation and velocity of an object. In traditional vehicle applications, the guidance system is providing destination navigation assistance along a route. Road level accuracy of a few meters is adequate. In a typical AV application, the INS works in conjunction with HD maps and perception systems to not only navigate a route but also for localization. Localization enables critical functions such as lane keeping and obstacle avoidance. Precise absolute position as well as relative position accuracy is important. This means you need lane level accuracy of under 10 centimeters. So when all systems are operating normally with nominal environmental conditions, good satellite coverage, an INS with a traditional automotive grade IMU provides sufficient positioning accuracy and reliability for safe operations. However, environmental conditions are often less than ideal. GPS signals are lost or degraded due to urban canyons, tunnels, overpasses, multi-path errors, poor satellite coverage. You may encounter conditions such as precipitation or reflective surfaces which can compromise the performance or integrity of data from camera, LiDAR and radar. In all cases, it's still imperative that the autonomous vehicle is able to reliably continue to navigate, safely maneuver to a stop and or request intervention. Fortunately, the one external factor we can always rely on to remain constant is Earth's gravity. Regardless of environmental conditions, the IMU will keep sensing and provide position data. Using the appropriate level of IMU performance can be the difference between making it through a tunnel or not. The task of navigating when perception sensor and GNSS data is unreliable is a process called dead reckoning. Here, the navigation system relies on other sensors, primarily the IMU, to continue to steer the vehicle safely. Menoms-based inertial sensors such as gyroscopes and accelerometers have long been used in vehicles as discrete safety components. For collision detection, airbag deployment and electronic stability control. High-end IMUs using MEMS or fiber optic technology are commonly utilized in aircraft and tactical guidance systems and they offer performance way higher than traditional MEMS sensors. As autonomous vehicle technology and safety standards progress, it's becoming apparent that the positioning accuracy and precision required from IMUs and INS is approaching those for aerospace and tactical grade devices. Consistent, reliable centimeter level accuracy rather than meter level accuracy. Until recently, this level of performance and safety integrity for IMUs has been cost prohibitive for high volume markets such as automotive. However, we are now seeing innovations in design and manufacturing certainly from companies like ST that make high performance IMU technology accessible for a wide range of autonomous and industrial use applications. ASINA leverages ST sensor and processor technology to deliver high performance autonomous navigation solutions such as the open IMU-330. This is a triple redundant, six-dough IMU, fully calibrated and stable over temperature. It's offered on ASINA's open source development platform. The triple redundancy enables fault tolerance for safety critical applications. The 1.5 degree per hour bias stability ensures higher accuracy for longer duration dead reckoning during GNSS coverage gaps. IMUs like this with built-in redundancy provide even higher precision estimates with further benefits of safety, integrity and reliability for the entire AV system and sensor fusion network. Another exciting trend in precise positioning is the emergence and expansion of RTK or real-time kinematics. RTK improves GPS positioning accuracy by more than a factor of 100 from meter level accuracy down to centimeter level. RTK technology refines the position data received from GPS signals by correcting for ionospheric and tropospheric delays, multi-path, satellite clock and ephemeral errors. Network RTK systems use survey grade base stations in a 50 kilometer radii which broadcast corrections wirelessly to rovers. The corrections are fused with GPS and IMU data through complex algorithms and common filters to provide a final position specific to the rover. Until recently RTK and services such as precise point positioning have come with a hefty price tag and long acquisition times. They have been used primarily in agriculture, land survey, construction applications for off-road vehicles in geofenced areas. The proliferation of autonomous technology and the need for precise positioning is given rise to new RTK solutions that are scalable across geographies and are economical, easy to integrate and optimized for AV sensor fusion. Cassina is excited to introduce our RTK positioning platform which includes the open RTK-330 hardware and our OpenARC streaming software. The OpenRTK-330 is a high-performance and cost-effective INS module powered by ST's Tessio 5 dual-band GNSS chipset. It's integrated with a triple redundant IMU like the IMU-330 and an RTK precise positioning engine. It has an open source framework which provides flexibility for developing your own customized applications. OpenARC is a proprietary GNSS correction network built to serve the needs of modern automotive and robotics applications that require precise performance, security and reliability while remaining affordable. Some more details about the OpenRTK-330 module. As mentioned, it integrates a GNSS receiver, high-performance IMU, GNSS corrections and a precise positioning engine with embedded INS sensor fusion algorithms. It includes an entry client and server for connecting to OpenARC or other streaming services as well as a web server. Peripherals are many including Ethernet, CANBUS, UART which can be used for development as well as final system integration. This product is powerful enough that it can be used out of the box as a standalone INS with OpenARC GNSS corrections and flexible enough that you can embed and customize your own algorithms and applications. Typically, it would be used on a rover but you can even program it to be a mini base station. The OpenRTK evaluation kit provides access to all the peripherals via connectors and also comes with an ST-Link debugger for development. Used along with Asina's open source development tools, it's a convenient platform to quickly simulate and develop your application as well as log, test and evaluate data. Asina's open source development platform works with our OpenIMU and OpenRTK hardware and includes online tools, IDE, GitHub and a community forum. For the development environment, there is an Asina extension for Visual Studio Code which includes all the published code and applications which are also available on GitHub. You can use these applications as they are or as a starting point for developing your own. Navigation Studio is an online GUI for viewing real-time data which can also be used to log and export data. Documentation and details for each product and application is available through our website where our community forum is also a great resource. Please visit our virtual booth at the conference for more resources. I'm also excited to share some real-time drive test results using the OpenRTK 330 INS. We used a Novitel span CPT unit for our reference to gather data and statistics on position accuracy. This drive consisted of open sky conditions with multiple overpasses. As you can see, with the OpenRTK you can achieve under 3 cm of horizontal position accuracy, 95% of the time and under 11 cm, 99% of the time using only GNSS, GNSS corrections and wheel odometry. While open sky conditions are ideal for receiving a reliable GNSS signals, tunnels are the opposite. This is where your IMU performance kicks in for dead reckoning. In this tunnel test, we again use Novitel as a reference and are also comparing the OpenRTK to another product on the market. Driving through a 1.5 km long tunnel which is just over 1.5 minutes of GNSS outage, we are able to maintain a position accuracy of under 3 m for most of the drive. The other solution, on the other hand, starts to drift almost immediately due to a less stable IMU. 3 m sounds like a lot, but keep in mind that this test is isolating the accuracy of the INS engine. Most automotive AV or ADAS applications would also be using perception for lane keeping and navigation. Having confidence in the IMU and positioning engine performance adds an extra level of reliability and safety for the overall system position accuracy. To wrap up, I would like to reiterate how essential it is for any autonomous vehicle to precisely know its location, its destination, and how to get there. The combined capabilities of advanced IMUs and RTK promise to democratize inertial navigation systems that provide the high performance, safety, and enhanced integrity that is vital for scalable and secure autonomous operation. Casino works with great companies like ST in striving to provide you with high performance solutions and platforms that are easy to use for development and system integration to make precise navigation easy. Thank you for your time. Please reach out if you have any questions and don't forget to visit our virtual booth for additional resources. Thank you and enjoy the rest of the conference.