Differential vectoring with 2 electric front motors

[hojbota]

I am building a trike, it has 2 front wheels and 1 to the rear.
The trike has a tilting function, it's recumbent with an enclosed cabin. Now for testing I'm preparing it for 1 motor to the rear wheel but the plan is to have 1 electric motor in each front wheel, to have better regen and using differential vectoring for acceleration and braking to add stability. The motors I'm planning to buy are the " grin technologies" all axle hub motor. Has anyone tried to do differential vectoring with hub motors. I'm sure it's possible but the question is if anyone has an idea of how hard it might be.
For more details about the project you can see it on YouTube, channel name hojbota-Ptv

 

[amberwolf]

There have been discussions about it, and at least one implementation; unfortunately the search is useless for finding them because the applicable terms are used for too many things and it doesn't do boolean.

If you use an FOC controller for each motor (VESC, Phaserunner, ASI, Fardriver, etc), then you have direct torque control over them, and your steering system can be used to vary the torque difference between wheels proportionally in turns via the throttle inputs.

 

[fechter]

I don’t know how much it would help. Independent front motors will give you sort of a differential effect that would be good for most conditions. At maximum turning angle one side might fight the other a little but I don’t think it would be very noticeable.

 

[Comrade]

That's an insanely high riding trike. Almost a penny-farthing of trikes. Why do you even want to do differential vectoring on something incapable of turning?

 

[hojbota]

That's an insanely high riding trike. Almost a penny-farthing of trikes. Why do you even want to do differential vectoring on something incapable of turning?

It tilts into corners like a bike

 

[hojbota]

There have been discussions about it, and at least one implementation; unfortunately the search is useless for finding them because the applicable terms are used for too many things and it doesn't do boolean.

If you use an FOC controller for each motor (VESC, Phaserunner, ASI, Fardriver, etc), then you have direct torque control over them, and your steering system can be used to vary the torque difference between wheels proportionally in turns via the throttle inputs.

I will do a system that have a gyro for inclination, throttle or brake input and steering angle. So for example I'm exiting a corner and accelerate, the acceleration to be distributed between the two front wheels based on the sensor data from steering, gyro and throttle. I will try first on an Arduino to see how it works

 

[hojbota]

That's an insanely high riding trike. Almost a penny-farthing of trikes. Why do you even want to do differential vectoring on something incapable of turning?

It tilts into corners like a bike so no problem with falling over. The problem I'm having now that I have to fix on the design is that the suspension at high lean angle limits steering. I know what the problem is and how to fix so in a few months it should ride even better