Recovering Kinetic Energy
Disk and drum brakes do nothing but to dissipate the vehicle's kinetic energy as heat in a short timespan through friction. Precious fuel is burnt to accelerate the vehicle, and then all that kinetic energy goes altogether to waste.
The idea of recovering some kinetic energy and store it onboard the vehicle does make some sense. However, it is not so easy to put into practice.
So far, the methods proposed are mechanical or electrical. Mechanical methods involve storing energy as a pressurized inert gas, or in a flywheel. Of course appropriate components are required to automatically connect and disconnect the energy recovery system from the crankshaft or wheel axles, or an hydraulic transmission in at least one design.
Electric systems use one or more motor/generator, again connected to the crankshaft or wheel axles, and an array of batteries or ultracapacitors for energy storage.
The KERS used this year by some F1 teams can be either based on a flywheel or batteries, but I am not aware of any teams using flywheels. The KERS is designed to provide a sort burst of maximum power; in fact this device needs a dedicated cooling system and adopts particular solutions to provide very high charge and discharge currents.
Hybrid cars, as the now-famous Prius, use an energy recovery system, but the main point of hybrids is a large battery pack allowing them to travel long distances on electric power alone - they are not designed for performance.
Energy recovery systems add mass to a vehicle and thus increase fuel consumption somewhat; then they are most useful when brakes are used frequently, on urban or winding roads.
Hybrid cars are enjoying some success, but there are legitimate doubts on their real energy efficiency. I think that a KERS could find some use in sport cars (maybe even as an aftermarket part?), while hydraulic or hydropneumatic systems are more suitable for heavy vehicles like trucks or buses.