Computational problems, Yes, the Solution of computational problems, that can never be solved in a feasible time period by the existing supercomputers is the basic need of a Quantum computer. Quantum Computers use quantum-mechanical phenomena like Superposition and Entanglement to perform computation. Superposition refers to the presence of a quantum particle like an atom or a photon in two states simultaneously, exactly it is not simultaneous but it spins so quickly that it appears to be in both positions at the same time. Entanglement is a physical phenomenon in which a group of quantum particles, interact, share spatial proximity in such a way that their quantum states remain dependent on one other even if they get separated by larger distance. Such particles show wave-like properties, and when frequency and waveform of these particles become identical, they are said to be in Coherence. Coherence is a mandatory condition to perform computing. In a Quantum System, Superposi...
What are Self Driving cars?
Self Driving cars are those vehicles that don't require human intervention to drive although human beings can drive such vehicles by using"Human Override Mode".
Autonomous cars are equipped with various sensors to map their environment and navigate appropriately.
 |
| Autonomous Vehicle |
Self Driving cars Technology
Sensors
Radar uses radio waves to determine the range, velocity, and angle of the objects moving around.
Lidar makes 3D images of the objects using laser light reflection.
Sonar uses sound wave propagation to navigate, sonar helps the control systems to identify different sounds coming from the objects in the surrounding.
GPS sensors help in traffic analysis location sensing.
Inertial Measurement System determines the body's specific force, angular rate, and orientation using Magnetometer, Accelerometers, and Gyroscopes.
All this sensory data is sent to the control systems of the vehicle where a Computer(Programmed with Deep Neural Network) analyses the data and performs the best possible response.
The Deep neural network takes care of SLAM, Simultaneous localization, and Mapping, a Computational geometry problem of "constructing or updating a map of an unknown environment while simultaneously keeping track of an agent's location within it."
How intelligent are Autonomous Cars?
The intelligence level of Cars is determined by their SAE ( Society of Automotive Engineers ) levels of driving automation ranging from 0 (fully manual) to 5 (fully autonomous).
SAE level 0 - Vehicles are fully manual.
SAE level 1("Hands-on") - Several systems help the driver but ultimate control is in driver's hands.
SAE level 2(" Hands-off ") - The automated system takes full control of the vehicle but the driver must monitor the driving and be prepared to intervene immediately at any time if the automated system fails to respond properly.
SAE level 3(" Eyes-off") - The driver can safely turn their attention away from the driving tasks but he must still be prepared to intervene as the vehicle system alerts him the orderly fashion for his turn.
SAE level 4("Mind-off") - No driver's attention is ever required for safety. Self-driving is supported only in limited spatial areas. The vehicle can safely park itself if the driver does not respond.
SAE level 5("Steering Wheel Optional") - No human intervention required.
Challenges faced by Self-driving cars
Collision Detection
Collision Detection is a Computational problem in detecting the intersection of two or more vehicles.
It can be a detection of 2D or 3D collision.
It's a challenge for a car to avoid collisions so they require lightning-fast response processing to mimic human reflexes.
That's why 5G technology is a must for Self-driving cars.
Object Detection
In order to move ahead, a car needs to know the objects (static and dynamic) and the surrounding environment.
The cars use Lidar to overcome this challenge.
Current Lidar systems use rotating hexagonal (6 sided) mirrors to split laser beam.
Three upper beams are used to map objects around the car and the other three are used to map the road and lanes.
Everything seems good about LIDAR in normal weather conditions but its not the same case in rainy or poor weather conditions because laser beams are partially reflected by rain droplets which adds noise to data.
Navigation on Unmapped Roads
Mapping of roads that have never been visited by the cars sometimes becomes a tough task as obstacles, traffic, weather, and the environment of such roads is unknown.
Comments