Skip to main content

Main body of the essay (Neural networks in the automotive sector)

Neural networks are being used in numerous commercial products in numerous industries. In the automotive industry neural networks have many potential applications both inside the vehicle, such as self-driving, advanced driving assistance systems (ADAS), and outside the vehicle such as during the development stage (Luckow et al.,2016).

There are mainly three types of neural networks which are mainly used in the automotive industry, recurrent neural networks, reinforcement learning NNs and convolutional NNs (Hernandez, 2017). A recurrent neural network (RNN) is a type of artificial neural network which uses sequential data or time series data (IBM,2021). In autonomous vehicles they are used to track moving objects and detect possible collisions, and with time RNNs can predict paths of moving objects, for example a moving pedestrian coming close to a road will probably mean that the pedestrian will cross it and the RNNs will make a decision on what the vehicle should do to avoid any danger, the other type of NNs are CNNs (Convolutional Neural Networks) which are used for visual pattern recognition and are used for obstacle detection in automotive world (Cadence PCB solutions, 2021). The best feature of CNNs and RNNs is their ability to increase accuracy over time using data, a self-driving car company Wayve believes that the two are the solution for self-driving cars to be truly successful, they claim that no matter how many programs a vehicle has, it will never be able to predict every situation, they argue that it is best to implement AI into vehicles to predict even in new situations that have not been programmed (ibid).

Here are a few use cases for neural networks in automotive industry, first is smart manufacturing  in which neural networks can help increase operational efficiency, additionally there is 'connected vehicles' which contains personal assistants and infotainment systems, furthermore and most importantly is the autonomous driving (NetApp, 2021).However, there are a few challenges for NNs in the automotive sector, while NNs empower vehicles to learn, it still lacks ethical reasoning that humans use, an example is a car choosing to hit multiple pedestrians or driving off a cliff and kill its occupants, such situations must be programmed by engineers and programmers, which make it very difficult to reach fully self-driving vehicles and creates pressure and hesitation which may continue for years (Cadence, 2021).


Comments

Post a Comment

Popular posts from this blog

Neural Network (Initial Idea 2)

Neural Networks   Neural Networks, colloquially known as artificial neural networks (ANNs) and also simulated neural networks (SNNs), are a sub-fiend of machine learning which in it self is a sub-field of Artificial Intelligence, they are the brain of AI and are components of deep learning. The name and structure of NNs are similar to the human brain and were inspired by it, they imitate the biological neurons by signaling to each other (IBM, 2020). ANNs consist of an input layer of neurons, one or two hidden layers of neurons and a final layer of output neurons (Wang, 2003).   The diagram illustrated exemplifies a simple architecture of a neural network, the lines which connect the neurons are called weights and are associated with a number which represent the strength of connection between neurons.   There are many different types of neural networks and each are used for specific motives. Below are some of the most common ones used in the field: The perceptron “t...
1 comment
Read more

Computer methodology - Artificial intelligence

Diagram of a simple feed-forward artificial neural network, with one “hidden layer,” also known as a “perceptron.” Image: Wikipedia Deep learning simply means “stacked neural networks” which are networks composed of several layers (Pathmind, 2021). Based on Rosenblatt’s perceptron, the simplest version of an artificial neural network has three layers of neurons. The first is the input layer which takes input values ,which in other words is data. This first layer of neurons is connected to the “hidden” layer. The outputs of these “hidden” neurons are then connected to the final output layer. This final layer is what gives you the answer to what the network has been trained to do (Medium, 2021).   The simplest neural networks consist of input layer, hidden layer and output layer, which are attached to each other with predictors, these predictors come attached with coefficients called "weights" , this is a non-linear network known as multilayer feed-foward network (Hyndman, R.J....
Post a Comment
Read more