Los Angeles is home to inVia Robotics, and if there’s one thing we know about here in Southern California, it’s traffic. Our freeways get so crowded that even a minor slowdown can cause a ripple effect that creates a major traffic jam, turning a 10-mile trip into a 2-hour crawl. Traffic planners try to reduce these backups by adding express lanes and using signals that regulate the flow of cars entering the freeway. The main problem with these measures is that cars are driven by people, and people don’t always follow the rules.
Think about driving up to a four-way intersection: You don’t know who’s coming from the other three directions, how fast they’re going, and whether they’ll turn or going straight. That’s why we have stop signs and traffic lights. And still, people routinely speed up to try and beat a red light or roll through a stop sign because they’re in a hurry, forcing you to slam on the brakes.
But what if you knew ahead of time how many cars were coming, how fast they were going, and which direction they were headed? You’d be able to sail through that intersection (and every other intersection) without stopping and get where you’re going in record time. That’s the ultimate goal for developers of self-driving cars and trucks: a nationwide network of autonomous vehicles, all synchronized to avoid collisions and slowdowns, carrying people and goods to their destinations.
Implementing such a system is still many years away, but it’s already a reality on a smaller scale — just not where you’d expect it to be.
Space-time motion planning allows us to synchronize our robots.
Autonomous Warehouse Robots Know the Way
The aisles of a warehouse are like the streets of a city, with straightaways and intersections providing paths to products. Send too many pickers down the same path and you’ve got a traffic jam. Autonomous warehouse robots — like the ones we build at inVia Robotics — have the advantage of synchronization. We’ve developed advanced space-time motion planning algorithms that allow us to synchronize the movement of our robots to avoid collisions and backups while simultaneously determining the most efficient route to their destinations. Instead of coming to a complete stop as they approach an intersection, our picking robots slow down fractionally to allow another to pass freely while maintaining forward momentum.
And those routes are continuously optimized via machine learning as orders come into the warehouse management system, which reduces pick time and increases picking accuracy to a rate of 99.9%.
Why Does it Matter?
Space-time motion planning gives you the green light to run your fulfillment operations at top speed without the fear of workflow interruptions, guaranteeing a steady stream of orders to your customers.