Autonomous driving is one of the world's most challenging computational problems. Very large amounts of data from cameras, RADARs, LIDARs, and HD-Maps must be processed to generate commands to control the car safely and comfortably in real-time. This challenging task requires a dedicated supercomputer that is energy-efficient and low-power, complex high-performance software, and breakthroughs in deep learning AI algorithms. To meet this task, the present technology provides advanced systems and methods that facilitate autonomous driving functionality, including a platform for autonomous driving Levels 3, 4, and/or 5. In preferred embodiments, the technology provides an end-to-end platform with a flexible architecture, including an architecture for autonomous vehicles that leverages computer vision and known ADAS techniques, providing diversity and redundancy, and meeting functional safety standards. The technology provides for a faster, more reliable, safer, energy-efficient and space-efficient System-on-a-Chip, which may be integrated into a flexible, expandable platform that enables a wide-range of autonomous vehicles, including cars, taxis, trucks, and buses, as well as watercraft and aircraft.

A control allocation system for a vehicle includes an electric power steering (EPS) system, one or more redundant actuation systems for controlling a plurality of wheels of the vehicle, and one or more controllers in electronic communication with the EPS system and the one or more redundant actuation systems. The one or more controllers execute instructions to determine tracking errors and vehicle dynamics states based on a plurality of local path planning references and receive a fault signal indicating the EPS system is non-functional. In response to receiving the fault signal, the one or more controllers determine a plurality of corrective constraints in real-time. The one or more controllers solve a real-time constrained optimization problem for each sampling interval of the control allocation system to determine a plurality of control actions based on the plurality of corrective constraints and the tracking errors.

A technique including capturing, by one or more cameras of a set of cameras disposed about a vehicle, one or more images, wherein a surround view system of the vehicle is configured to render a surround view image using a first hardware accelerator based on the one or more images, determining that a first hardware accelerator is unavailable, and rendering the surround view image using a second hardware accelerator based on the captured one or more images.

A vehicle power supply system includes a main power supply system including a main low-voltage power supply and a normal load; and a backup power supply system including a backup low-voltage power supply and an emergency important load. The backup power supply system includes a backup power supply control device. The backup power supply control device executes a suppliable electrical energy estimation process of estimating suppliable electrical energy suppliable from the backup low-voltage power supply to the emergency important load. The backup power supply control device outputs a signal based on a first electrical energy threshold value and a second electrical energy threshold value.

A vehicle power supply system includes a main power supply system including a main low-voltage power supply and a normal load; and a backup power supply system including a backup low-voltage power supply and an emergency important load. The backup power supply system includes a backup power supply control device. The backup power supply control device executes a suppliable electrical energy estimation process of estimating suppliable electrical energy suppliable from the backup low-voltage power supply to the emergency important load. The backup power supply control device outputs a signal based on a first electrical energy threshold value and a second electrical energy threshold value.

A control architecture for a vehicle communicatively connects each of a plurality of controllers to all of a plurality of commanded components. Each of the plurality of controllers is configured to send commands to be executed by one or more of the plurality of commanded components. Each of the plurality of commanded components determines that one of the plurality of controllers is a master controller on the basis of at least one signal transmitted from at least one of the plurality of controllers to the plurality of commanded components.

A control architecture for a vehicle communicatively connects each of a plurality of controllers to all of a plurality of commanded components. Each of the plurality of controllers is configured to send commands to be executed by one or more of the plurality of commanded components. Each of the plurality of commanded components determines that one of the plurality of controllers is a master controller on the basis of at least one signal transmitted from at least one of the plurality of controllers to the plurality of commanded components.

A redundant processing fabric in an autonomous vehicle may include: processing, by a first processing unit of a plurality of processing units, sensor data from a first sensor of a plurality of sensors, where the plurality of processing units are coupled to the plurality of sensors via a switched fabric, wherein the plurality of processing units and plurality of sensors are included in the autonomous vehicle, wherein the sensor data corresponds to an environment external to the autonomous vehicle; determining a failure in processing the sensor data by the first processing unit; and severing, in the switched fabric, a first communications path between the first sensor and the first processing unit; and establishing, in the switched fabric, a second communications path between the first sensor and a redundant processing unit.

A management domain unit is connected, via an independent virtual network (95) using a hypervisor unit, to a separate management domain unit (B) (10B) of a separate electronic control unit ECU (B) (1) having the separate management domain unit (B) (10B) and a separate basic domain unit (A) (50B) which substitutes for a basic domain unit. When an abnormality of the basic domain unit is detected, the management domain unit halts operation of the basic domain unit, and causes the separate management domain unit (B) (10B) possessed by the separate electronic control unit ECU (B) (1) to start operation of the separate basic domain unit (A) (50B).

A management domain unit is connected, via an independent virtual network (95) using a hypervisor unit, to a separate management domain unit (B) (10B) of a separate electronic control unit ECU (B) (1) having the separate management domain unit (B) (10B) and a separate basic domain unit (A) (50B) which substitutes for a basic domain unit. When an abnormality of the basic domain unit is detected, the management domain unit halts operation of the basic domain unit, and causes the separate management domain unit (B) (10B) possessed by the separate electronic control unit ECU (B) (1) to start operation of the separate basic domain unit (A) (50B).