How To Move Towards a Self-Driving Car Society with Electric Vehicles
OCTOBER 22 2023
This is a 20-30 year long-term vision.
Mix of human drivers with self-driving cars ❌
Entirely just self-driving cars ✅
Why? Efficiency increases significantly.
[https://youtu.be/_rVBHoBlmY4 How a self-driving cars society might look like, from one of my favourite TV shows "Upload".]
[https://youtu.be/iHzzSao6ypE A good video explaining traffic, and also why human driving will always be inefficient.]
Currently one can invest heavily in 5G/6G/(7G?) connectivity, IoT technologies and related research. This connectivity at all intersections will facilitate the rapid network needed.
Work to standardize self-driving car technology guidelines and regulations. Soon mandate new vehicles manufactured to atleast have the capability to upgrade to self-driving (in the form of digital remote control.)
This is also a reason to invest in electric vehicles! EVs are much easier to upgrade to self-driving considering that they are essentially an electronically controlled motor moving the vehicle, certainly much easier than digitally controlling a traditional gasoline engine vehicle.
Some more reasons to invest in EVs can be found in the subsequent blogpost!
Roll out region by region, create neighborhoods where only self-driving cars are allowed. Entering such a neighborhood automatically switches to self-driving mode with early warning and indications in route planners.
Expand until entire city is just a network of self-driving cars.
This is the future. Invest towards it.
Additional Note: While other places might take some time to catch up with making the roads and traffic more orderly, I believe Singapore will be one of the leaders in this area, with related research and progress already underway.
Why Invest in Electric Vehicles, Electric Motors, and Hybrids?
OCTOBER 7 2024
In history the first vehicles explored included steam, gasoline-powered models, and in fact electric as well early on itself. Though gasoline cars became dominant due to advancements in mass production and the higher energy density of gasoline. Today, with significant improvements in electricity infrastructure and battery technology, electric vehicles are becoming more practical and economically competitive, which will likely reduce the supply demand equilibrium for gasoline vehicles over time. One does not have to be reliant on the oil commodity trade and market dominators, as renewables can get you energy from anywhere on the planet.
If you don’t own a race car, electric vehicles provide insanely faster acceleration (and deceleration.) I have no words to describe the time when a I was in a Tesla EV with a friend and he hit the floor on the accelerator to show me the race car like acceleration 😵 While on a daily basis accelerating to top speeds may not be common, higher acceleration does help to move more efficiently through traffic, and more importantly rapid deceleration and braking is useful for preventing a lot of road accidents.
360° and omnidirectional wheels!! Convenient parking, more road space efficiency, and more. No non-electric motor gasoline vehicle will be doing this.[https://youtu.be/2EK3SFf5jWU Goodyear's Spherical Tires]
[https://youtu.be/D_eRkeAsMAE Hyundai Ioniq 5 Performs 360 Degree Turn and Crab Walk]EVs can have both the front and rear wheels providing torque (front and rear drive mixed systems) - This enhances mobility! For a situation where the back wheels are not able to get you out, the front wheels can help.
Overall EVs are more energy efficient. People argue on how electricity may still be produced with coal power plants, however a centralised power plant could manage it’s emissions better than individual vehicle exhausts on the road where gasoline/diesel vehicles are also a major cause of urban air and carbon monoxide pollution especially in cities with high population densities and inadequate management. Eventually the goal is renewable, and thus renewable energy is a simultaneous requirement needed. Lithium mining is resource intensive, and thus EVs may have higher initial emissions, though EVs are inherently more energy-efficient compared to internal combustion engine (ICE) vehicles, and thus over-time and over the EV lifespan the emissions are generally lower (especially if powered by renewables) than gasoline/diesel internal combustion engine ICE vehicles.
Much much easier to implement autonomous driving with EVs as mentioned in the previous blogpost How To Move Towards a Self-Driving Car Society.
Especially in urban areas with higher traffic, considering that the vehicle brake a lot, regenerative braking (retrieving energy from the motion of the car to recharge the battery) have higher efficiencies in EVs.
FYI “Hybrids”: Using a petrol/diesel engine to generate electricity, store it in a battery, and power electric motors.
Some more thoughts and ideas on the industry
Many people fear or distrust driverless vehicles, but my perspective is that this is largely in part due to that roads and vehicle signage around the world currently being human-centric for now, focussing on making clear and visible signs readable by humans (although this itself is abysmal and inadequate in some places) and thus the current environment by no means is feasible for driverless vehicles. However rather considering it as a trend requiring incremental investments and thinking about the ideal future goal, in a driverless world there’d be AI centric signage and communication. For a while to come, it would be an intermediary mix of the two.
For example, while driving an appropriate spacing between your vehicle and the vehicle in front of you is required, with more spacing required if behind a large super heavy vehicle, and this is something we’d understand from just our sight. From the 2D/2.5D camera perspective a driverless vehicle could evaluate the approximate size of the vehicle in the front, but would not know which class of vehicles it belongs to - super heavy vehicles, gasoline trucks, top open trucks, etc. that have different risks and spacing requirements and most definitely you don’t want to be caught behind a truck carrying a wind turbine or rocket part with your driverless vehicle incorrectly deeming it as an ordinary truck. Thus a system for vehicles to communicate with each other is needed for driverless vehicle to better understand it’s surrounding environment. This begins with standardization (and strictly enforcing it.)
In the future, 5G, IR signal or IOT could be used. An idea I had for an intermediary step could be enhancing and enforcing vehicle license plate legibility, and such that it is easily readable by a driverless vehicles (and possibly from any angle) akin to a QR code for humans. Additionally license plate number zoning could be useful where a certain range of numbers are issued only to a certain class of vehicles and further innovation in a digital and centralized national register of vehicle types, license plate information, issued date, etc that a driverless vehicle could reference.
Currently in INDIA the following kinds of vehicle number plate categorization is implemented. The proposal being to make changes to make it easily readable by driverless vehicles as well and adding more information to the number by categorization.
This kind a centralized national register or even number zoning would also be useful in curbing illegal license plate numbers that would be easily detectable by other sensors or driverless vehicles, aiding law enforcement.
Another idea, imagine a device that can be fitted onto existing vehicles, that would let them communicate with each other while providing vehicle dimensions, spatial orientation information, current velocity, etc for another device in the vicinity to quickly map out and predict imminent collisions, or detect collisions and prevent further collisions from other approaching vehicles by stopping or slowing down. Or automatic stops if receiving information on a traffic signal red light. Reliability issues would be a challenge, but it could start off as a warning and slowing down system.
On the matter of 5G technologies (which will enable driverless vehicles), should the goal of a city network engineer be to transition all communication to 5G technologies? Would a system using a mix of 5G and 4G be more beneficial? Do they both use the same part of the spectrum? What are the differences? What about 6G and beyond? These were some questions I had. From my understanding, a city network engineer should aim for a hybrid of 5G and 4G rather than transitioning entirely to 5G. Maintaining 4G is practical for broader coverage and weather conditions, while 5G is better suited for high data needs in dense areas, offering significantly higher speeds, lower latency, and improved capacity, albeit at a higher deployment cost. They don't use the same spectrum—4G operates in lower bands, while 5G also utilizes higher frequencies like millimeter waves for greater bandwidth. Future technologies like 6G may tap into terahertz bands, quantum communication, and AI integration, though each new generation faces challenges, indicating a saturation in how much traditional innovations alone can push further. Just an interesting note, in academia though there will be people working on 7G technology (or even beyond) already, which is absurd considering that 5G itself is yet to be widely adopted, but alas such is the nature of academia.
On the matter of electricity management, demand management is essential. Realizing that electric vehicles will inevitably increase demand, adequate infrastructure and investments into the electric grid is needed (this and electricity for AI data centres.) For example innovations in smart grids that communicate and automatically adjust charging rates when electricity demand is low and smart chargers that shut off and prevent over-charging. Without this the addition of EVs to the existing grid will only uncontrollably shift the energy demand duck curve leading to grid instability and reliance on peaking power plants. Additionally resources into battery management (from a broad perspective) and metals needed (such as lithium and rare earth metals) is needed. Marking battery age in a standardized format, analysing whether is is more economical to use lithium for energy storage instead of electric vehicles, depending on what has more demand in a region, and strictly enforcing lithium end-cycle management possibly by compelling manufacturers themselves to be end recyclers of their own battery products.
When people argue against electric it’s mostly due to the prohibitive costs to transition to and sustain an EV economy, and less due to inherent flaws. The vehicles we have today are technologies developed and refined over more than a century. If you want to reduce the number of vehicles to aid climate change action, capping or limiting sales of old or polluting vehicles could be a good immediate step, but at a significant economic loss considering economies and jobs like mechanics are trained to support the current infrastructure well. A balanced approach, seeing it rather as a direction and aiming to transition sustainably is needed, investing in ground infrastructure, discussion (like this blog post :D), and training/re-skilling, etc. needed to support an EV and driverless vehicle economy. While talking though it may be good to advocate and support for extremes (entirely electric, entirely driverless, etc) even if unrealistic as any small changes or funding at all in that direction can be beneficial, than a lax approach. As is with all things in life, and as frustrating as it is, a nuanced and balanced approach is needed, never black and white.
Driverless vehicles and EV technologies have significantly gotten better and continue to do so. With the technologies available, if it were possible to swiftly replace all vehicles, roads could be completely autonomous, considering that the technology is now not a bottleneck. Rather adoption rate is the bottleneck, and that it would be a mix between driver and driverless vehicles on the road for a while to come, where we can only slowly transition. And while entirely driverless may be ideal, if it’s really not possible, then at least the following goal will always remain: to increase coordination, reduce traffic and most importantly reduce accidents.
