Shaping the Future of GenAI in Transportation
December 6, 2023
Ben Reardon
Traffic pattern prediction software uses data analysis to forecast vehicle numbers and density on roads. Its goal is to improve traffic flow, reduce congestion, and suggest the best routes. GenAI, through real-time data processing, enables live updates and adjusts predictions according to current conditions for more accurate and dynamic traffic management.
Ensuring Safety through Transparency
AI has the potential to revolutionize the world, but only if people are willing to utilize it. 40% of Americans do not trust AI, and therefore, 40% of America’s market is inactive. While 60% of Americans will use AI in their daily lives, an additional 40% of America's population will greatly increase the scalability of AI within transportation technologies. Rightfully so, people shouldn’t fully trust AI in autonomous driving because if AI technology makes an error, crashes will occur, and humans will be harmed. In response to people's hesitation, companies are thoroughly testing their systems. One of the most efficient ways of testing is through AI modeling, where they train the system to react to real-world scenarios and edge cases. It is crucial to gain the public's trust; therefore, transparency about how AI operates and its decision making process should be available to the public. While it is understandable that skepticism toward AI in transportation is prevalent, working to eliminate doubt will pave the way for wider acceptance, utilization, and scalability of AI technologies in the transportation field.
Addressing Pain Points, Operational Costs, and Strategic Solutions
Pain Points
- Training: Autonomous vehicles require immense amounts of data to be capable of making decisions, and AI can efficiently process the necessary amounts of data.
- Safety: AI helps enhance safety by enabling vehicles to make split-second decisions and detect obstacles, pedestrians, and other vehicles more accurately than human drivers.
- Efficiency: Autonomous systems optimize routes and reduce traffic congestion, making transportation more efficient overall.
- Accessibility: Autonomous vehicles can offer mobility solutions for individuals who cannot drive due to disabilities or their age, increasing accessibility to transportation.
- Traffic Congestion: AI-powered autonomous driving systems can potentially mitigate traffic congestion by optimizing traffic flow and reducing unnecessary stops or delays.
Operational Cost
- Regulatory Challenges: Regulations and policies regarding autonomous vehicles vary across regions. Some areas have strict regulations that might hinder widespread deployment.
- Safety Concerns: Despite advancements, ensuring the absolute safety of autonomous vehicles remains a concern. Accidents involving autonomous vehicles can raise public distrust and regulatory scrutiny.
- Technological Limitations: AI systems might not yet be fully equipped to handle all scenarios on the road, especially in complex environments or under unpredictable conditions like extreme weather.
- Infrastructure Readiness: Autonomous vehicles often rely on high-quality infrastructure, including well-maintained roads and proper signage. Inadequate infrastructure can pose a significant challenge.
Strategies
- Advocacy and Collaboration: Engaging with policymakers and regulators to shape favorable regulations and standards that support the safe deployment of autonomous vehicles.
- Investment in R&D: Continuously investing in research and development to enhance AI algorithms, sensors, and other technologies crucial for autonomous driving.
- Testing and Validation: Rigorous testing and validation processes to ensure safety and reliability, simulating a wide range of scenarios that autonomous vehicles might encounter on the road.
- Partnerships and Alliances: Collaborating with infrastructure providers, technology companies, and other stakeholders to improve infrastructure readiness and develop complementary technologies.
Leading the Way: INRIX and Miovision
As different companies are attempting to enter into the space of using AI in traffic prediction and autonomous driving, they have many different factors to consider. These factors range from pain points to operational costs, and they’ve developed a magnitude of strategies to overcome these factors. The main pain points that companies are attempting to solve are AI model training, improving safety, increasing efficiency, and overall accessibility, and decreasing traffic congestion. Many of the operational costs that companies endure consist of regulatory challenges, safety concerns, technological limitations, and infrastructure readiness. In response, companies use a magnitude of strategies, such as consistently testing and validating models, investing in research and development, collaborating with policymakers and regulators, and forming partnerships and alliances.
Two companies that focus on traffic pattern recognition and prediction with the usage of AI are INRIX and Miovision.
INRIX has created a worldwide network of traffic analysis, allowing it to make predictions on ETA`s, traffic congestion, alternative routes, and more. INRIX has just received 70 million dollars from an investment by Morgan Stanley. With this money, they plan to face some of the pain points by collecting more data and improving their mobile devices, cameras, and sensors on roadways.
Miovision collects traffic data using Internet of Things sensors and devices to track, analyze, and optimize traffic lights. It utilizes AI systems to come to conclusions about different road types and locations based on the time of day. With this technology, they have received $120 million in funding from TELUS Ventures. With this, they plan on doubling their staff and improving their technology, hopefully allowing them to become the platform by which cities everywhere measure, manage, and optimize traffic.
