Skip to Main Content

How Does Multi-Policy Decision Making (MPDM) Work?

Our MPDM autonomy platform uses a constant stream of data to run thousands of simulations per second, allowing it to predict potential hazards and choose the safest course of action at every moment.

One of the biggest challenges to autonomous driving is safely handling the unexpected. Our MPDM autonomy platform uses a constant stream of data to run thousands of simulations per second, allowing it to predict potential hazards and choose the safest course of action at every moment. MPDM doesn’t just calculate what cars, pedestrians and other dynamic agents might do, but also how those actions might interact and change. This makes MPDM especially capable of handling the full range of possible driving conditions in the real world.

Video Transcription

Driving a car safely can be a challenge, especially on uncomplicated urban roads. At May Mobility, we're building an autonomous vehicle system to handle those challenges and deliver a better driving experience that's safer, greener, and more accessible. Meet our proprietary multi-policy decision-making system: MPDM for short. It's the right technology to make that better driving experience a reality today.

So, let's see how it works. You're riding along in a May Mobility autonomous vehicle, heading to a doctor's appointment. Hmm, we've got a sticky situation up ahead. There's a tight turn on a busy street, and that other car looks like it is going to try and pass that bicycle. The MPDM system uses advanced decision-making algorithms to predict how other road users will react to us and to each other.

Let's see how MPDM handles this situation. Everyone in this scenario has multiple things they can choose to do. In a fraction of a second, the MPDM system uses the onboard computer to simulate these possibilities, along with multiple hypothetical what-if scenarios, to determine possible outcomes. Each outcome is given a score based on safety and comfort. Now, let's consider all the scores and see which action would be best.

If we go normally, we might be fine, or we might not. Like, what if the bicycle starts to move to the left into the lane, perhaps to make a left turn? This score set is better. Even if the other car reacts to the bike and crosses into our lane, we still get a safe and comfortable outcome.

The MPDM system determines that the best thing to do is slow down a bit on the curve and move to the right of it, which is what a good human driver would have done. With MPDM Autonomous Technology, we're not just simulating many possible outcomes, we're giving our vehicles an imagination.

MPDM is revolutionizing the field of autonomous technology, creating a safer, better way to get around.

A large group of May Mobility team members pose outside their HQ in Ann Arbor, Michigan

Learn More About May Mobility

Meet the team making a difference in the world of autonomous vehicle technology and learn more about the core values that drive us.

  1. Meet May